Adding upstream version 14.2.21.upstream/14.2.21upstream

Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>

38 files changed, 8163 insertions, 0 deletions

diff --git a/src/boost/libs/graph_parallel/build/Jamfile.v2 b/src/boost/libs/graph_parallel/build/Jamfile.v2
new file mode 100644
index 00000000..5256fa78
--- /dev/null
+++ b/src/boost/libs/graph_parallel/build/Jamfile.v2
@@ -0,0 +1,43 @@
+# Copyright (c) 2002 Trustees of Indiana University
+#
+# Distributed under the Boost Software License, Version 1.0.
+# (See accompanying file LICENSE_1_0.txt or copy at
+# http://www.boost.org/LICENSE_1_0.txt)
+
+import mpi ;
+
+project boost/graph_parallel
+    : requirements <include>../src
+    : source-location ../src
+    ;
+
+local optional_sources ;
+local optional_reqs ;
+
+if [ mpi.configured ]
+{    
+  lib boost_graph_parallel
+    : mpi_process_group.cpp tag_allocator.cpp
+    : <library>../../mpi/build//boost_mpi
+      <library>/mpi//mpi [ mpi.extra-requirements ]
+      <define>BOOST_GRAPH_NO_LIB=1
+      <link>shared:<define>BOOST_GRAPH_DYN_LINK=1
+      # # Intel compiler ICEs if we turn optimization on
+      <toolset>intel-vc71-win-9.1:<optimization>off
+      # Without these flags, MSVC 7.1 crash
+      # User reports that VC++ 8 no longer has this problem
+      <toolset>msvc-7.1:<cxxflags>-GR-
+      <local-visibility>global
+    ;
+
+}
+else
+{
+  message boost_graph_parallel
+      : "warning: Graph library does not contain MPI-based parallel components."
+      : "note: to enable them, add \"using mpi ;\" to your user-config.jam."
+      : "note: to suppress this message, pass \"--without-graph_parallel\" to bjam."
+      ;
+}
+
+boost-install boost_graph_parallel ;
diff --git a/src/boost/libs/graph_parallel/example/Jamfile.v2 b/src/boost/libs/graph_parallel/example/Jamfile.v2
new file mode 100644
index 00000000..e8877757
--- /dev/null
+++ b/src/boost/libs/graph_parallel/example/Jamfile.v2
@@ -0,0 +1,14 @@
+# Copyright 2009 Vladimir Prus.
+#
+# Distributed under the Boost Software License, Version 1.0.
+#    (See accompanying file LICENSE_1_0.txt or copy at
+#          http://www.boost.org/LICENSE_1_0.txt)
+
+
+project : requirements <library>../build//boost_graph_parallel 
+    <library>../../system/build//boost_system
+    <library>../../mpi/build//boost_mpi
+    ;
+
+exe breadth_first_search : breadth_first_search.cpp ;
+exe dijkstra_shortest_paths : dijkstra_shortest_paths.cpp ;
diff --git a/src/boost/libs/graph_parallel/example/breadth_first_search.cpp b/src/boost/libs/graph_parallel/example/breadth_first_search.cpp
new file mode 100644
index 00000000..e1dbc985
--- /dev/null
+++ b/src/boost/libs/graph_parallel/example/breadth_first_search.cpp
@@ -0,0 +1,111 @@
+// Copyright (C) 2004-2008 The Trustees of Indiana University.
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+//  Authors: Douglas Gregor
+//           Andrew Lumsdaine
+
+// Example usage of breadth_first_search algorithm
+
+// Enable PBGL interfaces to BGL algorithms
+#include <boost/graph/use_mpi.hpp>
+
+// Communicate via MPI
+#include <boost/graph/distributed/mpi_process_group.hpp>
+
+// Breadth-first search algorithm
+#include <boost/graph/breadth_first_search.hpp>
+
+// Distributed adjacency list
+#include <boost/graph/distributed/adjacency_list.hpp>
+
+// METIS Input
+#include <boost/graph/metis.hpp>
+
+// Graphviz Output
+#include <boost/graph/distributed/graphviz.hpp>
+
+// For choose_min_reducer
+#include <boost/graph/distributed/distributed_graph_utility.hpp>
+
+// Standard Library includes
+#include <fstream>
+#include <string>
+
+#ifdef BOOST_NO_EXCEPTIONS
+void
+boost::throw_exception(std::exception const& ex)
+{
+    std::cout << ex.what() << std::endl;
+    abort();
+}
+#endif
+
+using namespace boost;
+using boost::graph::distributed::mpi_process_group;
+
+/* An undirected graph with distance values stored on the vertices. */
+typedef adjacency_list<vecS, distributedS<mpi_process_group, vecS>, undirectedS,
+                       /*Vertex properties=*/property<vertex_distance_t, std::size_t> >
+  Graph;
+
+int main(int argc, char* argv[])
+{
+  boost::mpi::environment env(argc,argv);
+
+  // Parse command-line options
+  const char* filename = "weighted_graph.gr";
+  if (argc > 1) filename = argv[1];
+
+  // Open the METIS input file
+  std::ifstream in(filename);
+  graph::metis_reader reader(in);
+
+  // Load the graph using the default distribution
+  Graph g(reader.begin(), reader.end(),
+          reader.num_vertices());
+
+  // Get vertex 0 in the graph
+  graph_traits<Graph>::vertex_descriptor start = vertex(0, g);
+
+  // Compute BFS levels from vertex 0
+  property_map<Graph, vertex_distance_t>::type distance =
+    get(vertex_distance, g);
+
+  // Initialize distances to infinity and set reduction operation to 'min'
+  BGL_FORALL_VERTICES(v, g, Graph) {
+    put(distance, v, (std::numeric_limits<std::size_t>::max)());
+  }
+  distance.set_reduce(boost::graph::distributed::choose_min_reducer<std::size_t>());
+
+  put(distance, start, 0);
+  breadth_first_search
+    (g, start,
+     visitor(make_bfs_visitor(record_distances(distance, on_tree_edge()))));
+
+  // Output a Graphviz DOT file
+  std::string outfile;
+
+  if (argc > 2)
+    outfile = argv[2];
+  else {
+    outfile = filename;
+    size_t i = outfile.rfind('.');
+    if (i != std::string::npos)
+      outfile.erase(outfile.begin() + i, outfile.end());
+    outfile += "-bfs.dot";
+  }
+
+  if (process_id(process_group(g)) == 0) {
+    std::cout << "Writing GraphViz output to " << outfile << "... ";
+    std::cout.flush();
+  }
+  write_graphviz(outfile, g,
+                 make_label_writer(distance));
+  if (process_id(process_group(g)) == 0)
+    std::cout << "Done." << std::endl;
+
+  return 0;
+}
diff --git a/src/boost/libs/graph_parallel/example/dijkstra_shortest_paths.cpp b/src/boost/libs/graph_parallel/example/dijkstra_shortest_paths.cpp
new file mode 100644
index 00000000..161271a4
--- /dev/null
+++ b/src/boost/libs/graph_parallel/example/dijkstra_shortest_paths.cpp
@@ -0,0 +1,98 @@
+// Copyright (C) 2004-2006 The Trustees of Indiana University.
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+//  Authors: Douglas Gregor
+//           Andrew Lumsdaine
+
+// Example usage of dijkstra_shortest_paths algorithm
+
+// Enable PBGL interfaces to BGL algorithms
+#include <boost/graph/use_mpi.hpp>
+
+// Communication via MPI
+#include <boost/graph/distributed/mpi_process_group.hpp>
+
+// Dijkstra's single-source shortest paths algorithm
+#include <boost/graph/dijkstra_shortest_paths.hpp>
+
+// Distributed adjacency list
+#include <boost/graph/distributed/adjacency_list.hpp>
+
+// METIS Input
+#include <boost/graph/metis.hpp>
+
+// Graphviz Output
+#include <boost/graph/distributed/graphviz.hpp>
+
+// Standard Library includes
+#include <fstream>
+#include <string>
+
+#ifdef BOOST_NO_EXCEPTIONS
+void
+boost::throw_exception(std::exception const& ex)
+{
+    std::cout << ex.what() << std::endl;
+    abort();
+}
+#endif
+
+using namespace boost;
+using boost::graph::distributed::mpi_process_group;
+
+/* An undirected, weighted graph with distance values stored on the
+   vertices. */
+typedef adjacency_list<vecS, distributedS<mpi_process_group, vecS>, undirectedS,
+                       /*Vertex properties=*/property<vertex_distance_t, float>,
+                       /*Edge properties=*/property<edge_weight_t, float> >
+  Graph;
+
+int main(int argc, char* argv[])
+{
+  boost::mpi::environment env(argc,argv);
+
+  // Parse command-line options
+  const char* filename = "weighted_graph.gr";
+  if (argc > 1) filename = argv[1];
+
+  // Open the METIS input file
+  std::ifstream in(filename);
+  graph::metis_reader reader(in);
+
+  // Load the graph using the default distribution
+  Graph g(reader.begin(), reader.end(), reader.weight_begin(),
+          reader.num_vertices());
+
+  // Get vertex 0 in the graph
+  graph_traits<Graph>::vertex_descriptor start = vertex(0, g);
+
+  // Compute shortest paths from vertex 0
+  dijkstra_shortest_paths(g, start,
+                          distance_map(get(vertex_distance, g)));
+
+  // Output a Graphviz DOT file
+  std::string outfile = filename;
+  if (argc > 2)
+    outfile = argv[2];
+  else {
+    size_t i = outfile.rfind('.');
+    if (i != std::string::npos)
+      outfile.erase(outfile.begin() + i, outfile.end());
+    outfile += "-dijkstra.dot";
+  }
+
+  if (process_id(process_group(g)) == 0) {
+    std::cout << "Writing GraphViz output to " << outfile << "... ";
+    std::cout.flush();
+  }
+  write_graphviz(outfile, g,
+                 make_label_writer(get(vertex_distance, g)),
+                 make_label_writer(get(edge_weight, g)));
+  if (process_id(process_group(g)) == 0)
+    std::cout << "Done." << std::endl;
+
+  return 0;
+}
diff --git a/src/boost/libs/graph_parallel/index.html b/src/boost/libs/graph_parallel/index.html
new file mode 100644
index 00000000..611797da
--- /dev/null
+++ b/src/boost/libs/graph_parallel/index.html
@@ -0,0 +1,20 @@
+<html>
+<!--
+     Copyright (c) 2004-2009 Trustees of Indiana University
+    
+     Distributed under the Boost Software License, Version 1.0.
+     (See accompanying file LICENSE_1_0.txt or copy at
+     http://www.boost.org/LICENSE_1_0.txt)
+  -->
+<head>
+<meta http-equiv="refresh" content="0; URL=doc/html/index.html">
+</head>
+<body>
+Automatic redirection failed, please go to
+<a href="doc/html/index.html">doc/html/index.html</a>.
+<hr>
+<p>Distributed under the Boost Software License, Version 1.0. (See accompanying 
+file <a href="../../LICENSE_1_0.txt">LICENSE_1_0.txt</a> or copy 
+at <a href="http://www.boost.org/LICENSE_1_0.txt">www.boost.org/LICENSE_1_0.txt</a>)</p>
+</body>
+</html>
diff --git a/src/boost/libs/graph_parallel/meta/libraries.json b/src/boost/libs/graph_parallel/meta/libraries.json
new file mode 100644
index 00000000..1cfb99a4
--- /dev/null
+++ b/src/boost/libs/graph_parallel/meta/libraries.json
@@ -0,0 +1,16 @@
+{
+    "key": "graph_parallel",
+    "name": "GraphParallel",
+    "authors": [
+        "Jeremy Siek, Doug Gregor, and a University of Notre Dame team."
+    ],
+    "description": "The PBGL graph interface and graph components are generic, in the same sense as the the Standard Template Library (STL).",
+    "category": [
+        "Algorithms",
+        "Containers",
+        "Iterators"
+    ],
+    "maintainers": [
+        "K. Noel Belcourt <kbelco -at- sandia.gov>"
+    ]
+}
diff --git a/src/boost/libs/graph_parallel/src/mpi_process_group.cpp b/src/boost/libs/graph_parallel/src/mpi_process_group.cpp
new file mode 100644
index 00000000..050b0f29
--- /dev/null
+++ b/src/boost/libs/graph_parallel/src/mpi_process_group.cpp
@@ -0,0 +1,1113 @@
+// Copyright (C) 2004-2006 The Trustees of Indiana University.
+// Copyright (C) 2007  Douglas Gregor  <doug.gregor@gmail.com>
+// Copyright (C) 2007  Matthias Troyer  <troyer@boost-consulting.com>
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+//  Authors: Douglas Gregor
+//           Andrew Lumsdaine
+//           Matthias Troyer
+#include <boost/assert.hpp>
+#include <boost/graph/use_mpi.hpp>
+#include <boost/graph/distributed/mpi_process_group.hpp>
+#include <boost/serialization/vector.hpp>
+#include <boost/mpi/environment.hpp>
+#include <boost/lexical_cast.hpp>
+#include <memory>
+#include <algorithm>
+
+//#define DEBUG 1
+
+
+//#define MAX_BATCHES 1500
+#define PREALLOCATE_BATCHES 250  
+// 500 is a better setting for PREALLOCATE_BATCHES if you're not using process 
+// subgroups and are building 64-bit binaries.  250 allows all the CTest
+// tests to pass in both 32 and 64-bit modes.  If you create multiple process 
+// groups with PREALLOCATE_BATCHES at a reasonable level in 32-bit mode you 
+// _will_ run out of memory and get "malloc failed" errors
+
+//#define NO_ISEND_BATCHES
+//#define NO_IMMEDIATE_PROCESSING
+//#define NO_SPLIT_BATCHES
+#define IRECV_BATCH
+
+// we cannot keep track of how many we received if we do not process them
+#ifdef NO_IMMEDIATE_PROCESSING
+#undef IRECV_BATCH
+#endif
+
+#ifdef DEBUG
+#  include <iostream>
+#endif // DEBUG
+
+namespace boost { namespace graph { namespace distributed {
+
+struct mpi_process_group::deallocate_block
+{
+  explicit deallocate_block(blocks_type* blocks) : blocks(blocks) { }
+
+  void operator()(int* block_num)
+  {
+    block_type* block = (*blocks)[*block_num];
+
+    // Mark this block as inactive
+    (*blocks)[*block_num] = 0;
+
+#ifdef DEBUG
+    fprintf(stderr, "Processor %i deallocated block #%i\n", 
+            boost::mpi::communicator().rank(), *block_num);
+#endif
+
+    // Delete the block and its block number
+    delete block_num;
+    delete block;
+  }
+
+private:
+  blocks_type* blocks;
+};
+
+mpi_process_group::impl::incoming_messages::incoming_messages()
+{
+  next_header.push_back(headers.begin());
+}
+
+mpi_process_group::impl::impl(std::size_t num_headers, std::size_t buffer_sz, 
+                              communicator_type parent_comm)
+  : comm(parent_comm, boost::mpi::comm_duplicate),
+    oob_reply_comm(parent_comm, boost::mpi::comm_duplicate),
+    allocated_tags(boost::mpi::environment::max_tag())
+{
+  max_sent=0;
+  max_received=0;
+  int n = comm.size();
+  outgoing.resize(n);
+  incoming.resize(n);
+  
+  // no synchronization stage means -1
+  // to keep the convention
+  synchronizing_stage.resize(n,-1); 
+  number_sent_batches.resize(n);
+  number_received_batches.resize(n);
+  trigger_context = trc_none;
+  processing_batches = 0;
+  
+  // Allocator a placeholder block "0"
+  blocks.push_back(new block_type);
+
+  synchronizing = false;
+
+  set_batch_size(num_headers,buffer_sz);
+  
+  for (int i = 0; i < n; ++i) {
+    incoming[i].next_header.front() = incoming[i].headers.end();
+    outgoing[i].buffer.reserve(batch_message_size);
+  }
+
+#ifdef PREALLOCATE_BATCHES
+  batch_pool.resize(PREALLOCATE_BATCHES);
+  for (std::size_t i = 0 ; i < batch_pool.size(); ++i) {
+    batch_pool[i].buffer.reserve(batch_message_size);
+    batch_pool[i].request=MPI_REQUEST_NULL;
+    free_batches.push(i);
+  }  
+#endif
+}
+
+void mpi_process_group::impl::set_batch_size(std::size_t header_num, std::size_t buffer_sz)
+{
+  batch_header_number = header_num;
+  batch_buffer_size = buffer_sz;
+
+  // determine batch message size by serializing the largest possible batch
+  outgoing_messages msg;
+  msg.headers.resize(batch_header_number);
+  msg.buffer.resize(batch_buffer_size);
+  boost::mpi::packed_oarchive oa(comm);
+  oa << const_cast<const outgoing_messages&>(msg);
+  batch_message_size = oa.size();
+}
+
+
+mpi_process_group::impl::~impl()
+{
+  // Delete the placeholder "0" block
+  delete blocks.front();
+  if (!boost::mpi::environment::finalized()) 
+  for (std::vector<MPI_Request>::iterator it=requests.begin(); 
+       it != requests.end();++it)
+    MPI_Cancel(&(*it));
+}
+
+namespace detail {
+// global batch handlers
+void handle_batch (mpi_process_group const& self, int source, int, 
+    mpi_process_group::outgoing_messages& batch,bool out_of_band)
+{
+#ifdef DEBUG
+  std::cerr << "Processing batch trigger\n";
+  std::cerr << "BATCH: " << process_id(self) << " <- " << source << " ("
+            << batch.headers.size() << " headers, "
+            << batch.buffer.size() << " bytes)"
+            << std::endl;
+#endif
+  // If we are not synchronizing, then this must be an early receive
+  trigger_receive_context old_context = self.impl_->trigger_context;
+  if (self.impl_->trigger_context != trc_in_synchronization)
+    self.impl_->trigger_context = trc_early_receive;
+
+  // Receive the batched messages
+  self.receive_batch(source,batch);
+
+  // Restore the previous context
+  self.impl_->trigger_context = old_context;
+}
+
+// synchronization handler
+void handle_sync (mpi_process_group const& self, int source, int tag, int val,
+                  bool)
+{
+  // increment the stage for the source
+  std::size_t stage = static_cast<std::size_t>(
+    ++self.impl_->synchronizing_stage[source]);
+
+  BOOST_ASSERT(source != process_id(self));
+
+#ifdef DEBUG
+  std::ostringstream out;
+  out << process_id(self) << ": handle_sync from " << source 
+      << " (stage = " << self.impl_->synchronizing_stage[source] << ")\n";
+  std::cerr << out.str();
+#endif
+
+  // record how many still have messages to be sent
+  if (self.impl_->synchronizing_unfinished.size()<=stage) {
+    BOOST_ASSERT(self.impl_->synchronizing_unfinished.size() == stage);
+    self.impl_->synchronizing_unfinished.push_back(val >= 0 ? 1 : 0);
+  }
+  else
+    self.impl_->synchronizing_unfinished[stage]+=(val >= 0 ? 1 : 0);
+
+  // record how many are in that stage
+  if (self.impl_->processors_synchronizing_stage.size()<=stage) {
+    BOOST_ASSERT(self.impl_->processors_synchronizing_stage.size() == stage);
+    self.impl_->processors_synchronizing_stage.push_back(1);
+  }
+  else
+    ++self.impl_->processors_synchronizing_stage[stage];
+    
+  // subtract how many batches we were supposed to receive
+  if (val>0)
+    self.impl_->number_received_batches[source] -= val;
+}
+
+
+}
+
+mpi_process_group::mpi_process_group(communicator_type parent_comm)
+{
+  // 64K messages and 1MB buffer turned out to be a reasonable choice
+  impl_.reset(new impl(64*1024,1024*1024,parent_comm));
+#ifndef IRECV_BATCH
+  global_trigger<outgoing_messages>(msg_batch,&detail::handle_batch);
+#else // use irecv version by providing a maximum buffer size
+  global_trigger<outgoing_messages>(msg_batch,&detail::handle_batch,
+         impl_->batch_message_size);
+#endif
+  global_trigger<outgoing_messages>(msg_large_batch,&detail::handle_batch);
+  global_trigger<int>(msg_synchronizing,&detail::handle_sync);
+  rank = impl_->comm.rank();
+  size = impl_->comm.size();
+  
+#ifdef SEND_OOB_BSEND
+  // let us try with a default bufferr size of 16 MB
+  if (!message_buffer_size())
+    set_message_buffer_size(16*1024*1024);
+#endif
+}
+
+mpi_process_group::mpi_process_group(std::size_t h, std::size_t sz, 
+                                     communicator_type parent_comm)
+{
+  impl_.reset(new impl(h,sz,parent_comm));
+#ifndef IRECV_BATCH
+  global_trigger<outgoing_messages>(msg_batch,&detail::handle_batch);
+#else // use irecv version by providing a maximum buffer size
+  global_trigger<outgoing_messages>(msg_batch,&detail::handle_batch,
+         impl_->batch_message_size);
+#endif
+  global_trigger<outgoing_messages>(msg_large_batch,&detail::handle_batch);
+  global_trigger<int>(msg_synchronizing,&detail::handle_sync);
+  rank = impl_->comm.rank();
+  size = impl_->comm.size();
+#ifdef SEND_OOB_BSEND
+  // let us try with a default bufferr size of 16 MB
+  if (!message_buffer_size())
+    set_message_buffer_size(16*1024*1024);
+#endif
+}
+
+mpi_process_group::mpi_process_group(const mpi_process_group& other,
+                                     const receiver_type& handler, bool)
+  : impl_(other.impl_)
+{ 
+  rank = impl_->comm.rank();
+  size = impl_->comm.size();
+  replace_handler(handler); 
+}
+
+mpi_process_group::mpi_process_group(const mpi_process_group& other,
+                                     attach_distributed_object, bool)
+  : impl_(other.impl_)
+{ 
+  rank = impl_->comm.rank();
+  size = impl_->comm.size();
+  allocate_block();
+
+  for (std::size_t i = 0; i < impl_->incoming.size(); ++i) {
+    if (my_block_number() >= (int)impl_->incoming[i].next_header.size()) {
+      impl_->incoming[i].next_header
+        .push_back(impl_->incoming[i].headers.begin());
+    } else {
+      impl_->incoming[i].next_header[my_block_number()] =
+        impl_->incoming[i].headers.begin();
+    }
+
+#ifdef DEBUG
+    if (process_id(*this) == 0) {
+      std::cerr << "Allocated tag block " << my_block_number() << std::endl;
+    }
+#endif
+  }
+}
+
+mpi_process_group::~mpi_process_group()  {
+  /*
+  std::string msg = boost::lexical_cast<std::string>(process_id(*this)) + " " +
+            boost::lexical_cast<std::string>(impl_->max_received) + " " +
+            boost::lexical_cast<std::string>(impl_->max_sent) + "\n";
+     std::cerr << msg << "\n";
+     */
+}
+
+
+mpi_process_group::communicator_type communicator(const mpi_process_group& pg)
+{ return pg.impl_->comm; }
+
+void
+mpi_process_group::replace_handler(const receiver_type& handler,
+                                   bool out_of_band_receive)
+{
+  make_distributed_object();
+
+  // Attach the receive handler
+  impl_->blocks[my_block_number()]->on_receive = handler;
+}
+
+void
+mpi_process_group::make_distributed_object()
+{
+  if (my_block_number() == 0) {
+    allocate_block();
+
+    for (std::size_t i = 0; i < impl_->incoming.size(); ++i) {
+      if (my_block_number() >= (int)impl_->incoming[i].next_header.size()) {
+        impl_->incoming[i].next_header
+          .push_back(impl_->incoming[i].headers.begin());
+      } else {
+        impl_->incoming[i].next_header[my_block_number()] =
+          impl_->incoming[i].headers.begin();
+      }
+
+#ifdef DEBUG
+      if (process_id(*this) == 0) {
+        std::cerr << "Allocated tag block " << my_block_number() << std::endl;
+      }
+#endif
+    }
+  } else {
+    // Clear out the existing triggers
+    std::vector<shared_ptr<trigger_base> >()
+      .swap(impl_->blocks[my_block_number()]->triggers); 
+  }
+
+  // Clear out the receive handler
+  impl_->blocks[my_block_number()]->on_receive = 0;
+}
+
+void
+mpi_process_group::
+replace_on_synchronize_handler(const on_synchronize_event_type& handler)
+{
+  if (my_block_number() > 0)
+    impl_->blocks[my_block_number()]->on_synchronize = handler;
+}
+
+int mpi_process_group::allocate_block(bool out_of_band_receive)
+{
+  BOOST_ASSERT(!block_num);
+  block_iterator i = impl_->blocks.begin();
+  while (i != impl_->blocks.end() && *i) ++i;
+
+  if (i == impl_->blocks.end()) {
+    impl_->blocks.push_back(new block_type());
+    i = impl_->blocks.end() - 1;
+  } else {
+    *i = new block_type();
+  }
+
+  block_num.reset(new int(i - impl_->blocks.begin()),
+                  deallocate_block(&impl_->blocks));
+
+#ifdef DEBUG
+  fprintf(stderr,
+          "Processor %i allocated block #%i\n", process_id(*this), *block_num);
+#endif
+
+  return *block_num;
+}
+
+bool mpi_process_group::maybe_emit_receive(int process, int encoded_tag) const
+{
+  std::pair<int, int> decoded = decode_tag(encoded_tag);
+
+  BOOST_ASSERT (decoded.first < static_cast<int>(impl_->blocks.size()));
+
+  block_type* block = impl_->blocks[decoded.first];
+  if (!block) {
+    std::cerr << "Received message from process " << process << " with tag "
+              << decoded.second << " for non-active block "
+              << decoded.first << std::endl;
+    std::cerr << "Active blocks are: ";
+    for (std::size_t i = 0; i < impl_->blocks.size(); ++i)
+      if (impl_->blocks[i])
+        std::cerr << i << ' ';
+    std::cerr << std::endl;
+    BOOST_ASSERT(block);
+  }
+  
+  if (decoded.second < static_cast<int>(block->triggers.size())
+      && block->triggers[decoded.second]) {
+    // We have a trigger for this message; use it
+    trigger_receive_context old_context = impl_->trigger_context;
+    impl_->trigger_context = trc_out_of_band;
+    block->triggers[decoded.second]->receive(*this, process, decoded.second, 
+                                             impl_->trigger_context, 
+                                             decoded.first);
+    impl_->trigger_context = old_context;
+  }
+  else
+    return false;
+  // We receives the message above
+  return true;
+}
+
+bool mpi_process_group::emit_receive(int process, int encoded_tag) const
+{
+  std::pair<int, int> decoded = decode_tag(encoded_tag);
+
+  if (decoded.first >= static_cast<int>(impl_->blocks.size()))
+    // This must be an out-of-band message destined for
+    // send_oob_with_reply; ignore it.
+    return false;
+
+  // Find the block that will receive this message
+  block_type* block = impl_->blocks[decoded.first];
+  BOOST_ASSERT(block);
+  if (decoded.second < static_cast<int>(block->triggers.size())
+      && block->triggers[decoded.second])
+    // We have a trigger for this message; use it
+    block->triggers[decoded.second]->receive(*this,process, decoded.second, 
+                                             impl_->trigger_context);
+  else if (block->on_receive)
+    // Fall back to the normal message handler
+    block->on_receive(process, decoded.second);
+  else
+    // There was no handler for this message
+    return false;
+
+  // The message was handled above
+  return true;
+}
+
+void mpi_process_group::emit_on_synchronize() const
+{
+  for (block_iterator i = impl_->blocks.begin(); i != impl_->blocks.end(); ++i)
+    if (*i && (*i)->on_synchronize) (*i)->on_synchronize();
+}
+
+
+optional<std::pair<mpi_process_group::process_id_type, int> >
+mpi_process_group::probe() const
+{
+#ifdef DEBUG
+  std::cerr << "PROBE: " << process_id(*this) << ", tag block = "
+            << my_block_number() << std::endl;
+#endif
+
+  typedef std::pair<process_id_type, int> result_type;
+
+  int tag_block = my_block_number();
+
+  for (std::size_t source = 0; source < impl_->incoming.size(); ++source) {
+    impl::incoming_messages& incoming = impl_->incoming[source];
+    std::vector<impl::message_header>::iterator& i =
+      incoming.next_header[tag_block];
+    std::vector<impl::message_header>::iterator end =  incoming.headers.end();
+
+    while (i != end) {
+      if (i->tag != -1 && decode_tag(i->tag).first == my_block_number()) {
+#ifdef DEBUG
+        std::cerr << "PROBE: " << process_id(*this) << " <- " << source
+                  << ", block = " << my_block_number() << ", tag = "
+                  << decode_tag(i->tag).second << ", bytes = " << i->bytes
+                  << std::endl;
+#endif
+        return result_type(source, decode_tag(i->tag).second);
+      }
+      ++i;
+    }
+  }
+  return optional<result_type>();
+}
+
+void
+mpi_process_group::maybe_send_batch(process_id_type dest) const
+{
+#ifndef NO_SPLIT_BATCHES  
+  impl::outgoing_messages& outgoing = impl_->outgoing[dest];
+  if (outgoing.buffer.size() >= impl_->batch_buffer_size ||
+      outgoing.headers.size() >= impl_->batch_header_number) {
+    // we are full and need to send
+    outgoing_messages batch;
+    batch.buffer.reserve(impl_->batch_buffer_size);
+    batch.swap(outgoing);
+    if (batch.buffer.size() >= impl_->batch_buffer_size 
+         && batch.headers.size()>1 ) {
+      // we are too large, keep the last message in the outgoing buffer
+      std::copy(batch.buffer.begin()+batch.headers.back().offset,
+                batch.buffer.end(),std::back_inserter(outgoing.buffer));
+      batch.buffer.resize(batch.headers.back().offset);
+      outgoing.headers.push_back(batch.headers.back());
+      batch.headers.pop_back();
+      outgoing.headers.front().offset=0;
+    }
+    send_batch(dest,batch);
+  }
+#endif
+}
+
+void
+mpi_process_group::send_batch(process_id_type dest) const
+{
+  impl::outgoing_messages& outgoing = impl_->outgoing[dest];
+  if (outgoing.headers.size()) {
+    // need to copy to avoid race conditions
+    outgoing_messages batch;
+    batch.buffer.reserve(impl_->batch_buffer_size);
+    batch.swap(outgoing); 
+    send_batch(dest,batch);
+  }
+}
+
+
+void
+mpi_process_group::send_batch(process_id_type dest, 
+                              outgoing_messages& outgoing) const
+{
+  impl_->free_sent_batches();
+  process_id_type id = process_id(*this);
+
+  // clear the batch
+#ifdef DEBUG
+  std::cerr << "Sending batch: " << id << " -> "  << dest << std::endl;
+#endif
+  // we increment the number of batches sent
+  ++impl_->number_sent_batches[dest];
+  // and send the batch
+  BOOST_ASSERT(outgoing.headers.size() <= impl_->batch_header_number);
+  if (id != dest) {
+#ifdef NO_ISEND_BATCHES
+    impl::batch_request req;
+#else
+#ifdef PREALLOCATE_BATCHES
+    while (impl_->free_batches.empty()) {
+      impl_->free_sent_batches();
+      poll();
+    }
+    impl::batch_request& req = impl_->batch_pool[impl_->free_batches.top()];
+    impl_->free_batches.pop();
+#else
+    impl_->sent_batches.push_back(impl::batch_request());
+    impl::batch_request& req = impl_->sent_batches.back();
+#endif
+#endif
+    boost::mpi::packed_oarchive oa(impl_->comm,req.buffer);
+    oa << outgoing;
+
+    int tag = msg_batch;
+    
+#ifdef IRECV_BATCH
+    if (oa.size() > impl_->batch_message_size)
+      tag = msg_large_batch;
+#endif
+
+#ifndef NDEBUG // Prevent uninitialized variable warning with NDEBUG is on
+    int result =
+#endif // !NDEBUG
+      MPI_Isend(const_cast<void*>(oa.address()), oa.size(),
+                MPI_PACKED, dest, tag, impl_->comm,
+                &req.request);
+    BOOST_ASSERT(result == MPI_SUCCESS);
+    impl_->max_sent = (std::max)(impl_->max_sent,impl_->sent_batches.size());
+#ifdef NO_ISEND_BATCHES
+    int done=0;
+    do {                                                        
+        poll();                                                
+        MPI_Test(&req.request,&done,MPI_STATUS_IGNORE);               
+       } while (!done);                                            
+#else
+#ifdef MAX_BATCHES
+    while (impl_->sent_batches.size() >= MAX_BATCHES-1) {
+      impl_->free_sent_batches();
+      poll();
+    }
+#endif
+#endif
+  }
+  else
+    receive_batch(id,outgoing);
+}
+
+void mpi_process_group::process_batch(int source) const
+{
+  bool processing_from_queue = !impl_->new_batches.empty();
+  impl_->processing_batches++;
+  typedef std::vector<impl::message_header>::iterator iterator;
+
+  impl::incoming_messages& incoming = impl_->incoming[source];
+  
+  // Set up the iterators pointing to the next header in each block
+  for (std::size_t i = 0; i < incoming.next_header.size(); ++i)
+        incoming.next_header[i] = incoming.headers.begin();
+
+  buffer_type remaining_buffer;
+  std::vector<impl::message_header> remaining_headers;
+
+  iterator end = incoming.headers.end();
+
+  for (iterator i = incoming.headers.begin(); i != end; ++i) {
+    // This this message has already been received, skip it
+    if (i->tag == -1)
+      continue;
+
+#ifdef BATCH_DEBUG
+    std::cerr << process_id(*this) << ": emit_receive(" << source << ", "
+              << decode_tag(i->tag).first << ":" << decode_tag(i->tag).second
+              << ")\n";
+#endif
+
+    if (!emit_receive(source, i->tag)) {
+#ifdef BATCH_DEBUG
+      std::cerr << process_id(*this) << ": keeping message # " 
+            << remaining_headers.size() << " from " << source << " ("
+            << decode_tag(i->tag).first << ":" 
+            << decode_tag(i->tag).second << ", " 
+            << i->bytes << " bytes)\n";
+#endif
+      // Hold on to this message until the next stage
+      remaining_headers.push_back(*i);
+      remaining_headers.back().offset = remaining_buffer.size();
+      remaining_buffer.insert(remaining_buffer.end(),
+                  &incoming.buffer[i->offset],
+                  &incoming.buffer[i->offset] + i->bytes);
+    }
+  }
+
+  // Swap the remaining messages into the "incoming" set.
+  incoming.headers.swap(remaining_headers);
+  incoming.buffer.swap(remaining_buffer);
+
+  // Set up the iterators pointing to the next header in each block
+  for (std::size_t i = 0; i < incoming.next_header.size(); ++i)
+    incoming.next_header[i] = incoming.headers.begin();
+  impl_->processing_batches--;
+  
+  if (!processing_from_queue)
+    while (!impl_->new_batches.empty()) {
+      receive_batch(impl_->new_batches.front().first,
+                    impl_->new_batches.front().second);
+      impl_->new_batches.pop();
+    }
+}
+
+
+void mpi_process_group::receive_batch(process_id_type source, 
+  outgoing_messages& new_messages) const
+{
+  impl_->free_sent_batches();
+  if(!impl_->processing_batches) {
+    // increase the number of received batches
+    ++impl_->number_received_batches[source];
+    // and receive the batch
+    impl::incoming_messages& incoming = impl_->incoming[source];
+    typedef std::vector<impl::message_header>::iterator iterator;
+    iterator end = new_messages.headers.end();
+    for (iterator i = new_messages.headers.begin(); i != end; ++i) {
+      incoming.headers.push_back(*i);
+      incoming.headers.back().offset = incoming.buffer.size();
+      incoming.buffer.insert(incoming.buffer.end(),
+                  &new_messages.buffer[i->offset],
+                  &new_messages.buffer[i->offset] + i->bytes);
+    }
+    // Set up the iterators pointing to the next header in each block
+    for (std::size_t i = 0; i < incoming.next_header.size(); ++i)
+      incoming.next_header[i] = incoming.headers.begin();
+#ifndef NO_IMMEDIATE_PROCESSING
+    process_batch(source);
+#endif
+  }
+  else {
+  #ifdef DEBUG
+    std::cerr << "Pushing incoming message batch onto queue since we are already processing a batch.\n";
+  #endif
+    // use swap to avoid copying
+    impl_->new_batches.push(std::make_pair(int(source),outgoing_messages()));
+    impl_->new_batches.back().second.swap(new_messages);
+    impl_->max_received = (std::max)(impl_->max_received,impl_->new_batches.size());
+  }
+}
+
+
+void mpi_process_group::pack_headers() const 
+{
+  for (process_id_type other = 0; other < num_processes(*this); ++other) {
+    typedef std::vector<impl::message_header>::iterator iterator;
+
+    impl::incoming_messages& incoming = impl_->incoming[other];
+
+    buffer_type remaining_buffer;
+    std::vector<impl::message_header> remaining_headers;
+
+    iterator end = incoming.headers.end();
+    for (iterator i = incoming.headers.begin(); i != end; ++i) {
+      if (i->tag == -1)
+        continue;
+
+      // Hold on to this message until the next stage
+      remaining_headers.push_back(*i);
+      remaining_headers.back().offset = remaining_buffer.size();
+      remaining_buffer.insert(remaining_buffer.end(),
+                              &incoming.buffer[i->offset],
+                              &incoming.buffer[i->offset] + i->bytes);
+    }
+    
+    // Swap the remaining messages into the "incoming" set.
+    incoming.headers.swap(remaining_headers);
+    incoming.buffer.swap(remaining_buffer);
+
+    // Set up the iterators pointing to the next header in each block
+    for (std::size_t i = 0; i < incoming.next_header.size(); ++i)
+      incoming.next_header[i] = incoming.headers.begin();
+  }
+}
+
+void mpi_process_group::receive_batch(boost::mpi::status& status) const
+{
+  //std::cerr << "Handling batch\n";
+  outgoing_messages batch;
+  //impl_->comm.recv(status.source(),status.tag(),batch);
+
+  //receive_oob(*this,status.source(),msg_batch,batch);
+    
+  // Determine how big the receive buffer should be
+#if BOOST_VERSION >= 103600
+  int size = status.count<boost::mpi::packed>().get();
+#else
+  int size;
+  MPI_Status mpi_status(status);
+  MPI_Get_count(&mpi_status, MPI_PACKED, &size);
+#endif
+
+  // Allocate the receive buffer
+  boost::mpi::packed_iarchive in(impl_->comm,size);
+  
+  // Receive the message data
+  MPI_Recv(in.address(), size, MPI_PACKED,
+           status.source(), status.tag(), 
+           impl_->comm, MPI_STATUS_IGNORE);
+
+  // Unpack the message data
+  in >> batch;
+  receive_batch(status.source(),batch);
+}
+
+std::pair<boost::mpi::communicator, int> 
+mpi_process_group::actual_communicator_and_tag(int tag, int block) const
+{
+  if (tag >= max_tags * static_cast<int>(impl_->blocks.size()))
+    // Use the out-of-band reply communicator
+    return std::make_pair(impl_->oob_reply_comm, tag);
+  else
+    // Use the normal communicator and translate the tag
+    return std::make_pair(impl_->comm, 
+                          encode_tag(block == -1? my_block_number() : block,
+                                     tag));
+}
+
+
+void mpi_process_group::synchronize() const
+{
+  // Don't synchronize if we've already finished
+  if (boost::mpi::environment::finalized()) 
+    return;
+
+#ifdef DEBUG
+  std::cerr << "SYNC: " << process_id(*this) << std::endl;
+#endif
+
+  emit_on_synchronize();
+
+  process_id_type id = process_id(*this);     // Our rank
+  process_size_type p = num_processes(*this); // The number of processes
+
+  // Pack the remaining incoming messages into the beginning of the
+  // buffers, so that we can receive new messages in this
+  // synchronization step without losing those messages that have not
+  // yet been received.
+  pack_headers();
+
+  impl_->synchronizing_stage[id] = -1;
+  int stage=-1;
+  bool no_new_messages = false;
+  while (true) {
+      ++stage;
+#ifdef DEBUG
+      std::cerr << "SYNC: " << id << " starting stage " << (stage+1) << ".\n";
+#endif
+
+      // Tell everyone that we are synchronizing. Note: we use MPI_Isend since 
+      // we absolutely cannot have any of these operations blocking.
+      
+      // increment the stage for the source
+       ++impl_->synchronizing_stage[id];
+       if (impl_->synchronizing_stage[id] != stage)
+         std::cerr << "Expected stage " << stage << ", got " << impl_->synchronizing_stage[id] << std::endl;
+       BOOST_ASSERT(impl_->synchronizing_stage[id]==stage);
+      // record how many still have messages to be sent
+      if (static_cast<int>(impl_->synchronizing_unfinished.size())<=stage) {
+        BOOST_ASSERT(static_cast<int>(impl_->synchronizing_unfinished.size()) == stage);
+        impl_->synchronizing_unfinished.push_back(no_new_messages ? 0 : 1);
+      }
+      else
+        impl_->synchronizing_unfinished[stage]+=(no_new_messages ? 0 : 1);
+
+      // record how many are in that stage
+      if (static_cast<int>(impl_->processors_synchronizing_stage.size())<=stage) {
+        BOOST_ASSERT(static_cast<int>(impl_->processors_synchronizing_stage.size()) == stage);
+        impl_->processors_synchronizing_stage.push_back(1);
+      }
+      else
+        ++impl_->processors_synchronizing_stage[stage];
+
+      impl_->synchronizing = true;
+
+      for (int dest = 0; dest < p; ++dest) {
+        int sync_message = no_new_messages ? -1 : impl_->number_sent_batches[dest];
+        if (dest != id) {
+          impl_->number_sent_batches[dest]=0;       
+          MPI_Request request;
+          MPI_Isend(&sync_message, 1, MPI_INT, dest, msg_synchronizing, impl_->comm,&request);
+          int done=0;
+          do {
+            poll();
+            MPI_Test(&request,&done,MPI_STATUS_IGNORE);
+          } while (!done);
+        }
+        else { // need to subtract how many messages I should have received
+          impl_->number_received_batches[id] -=impl_->number_sent_batches[id];
+          impl_->number_sent_batches[id]=0;
+        }
+      }
+
+      // Keep handling out-of-band messages until everyone has gotten
+      // to this point.
+      while (impl_->processors_synchronizing_stage[stage] <p) {
+        // with the trigger based solution we cannot easily pass true here 
+        poll(/*wait=*/false, -1, true);
+
+      }
+
+      // check that everyone is at least here
+      for (int source=0; source<p ; ++source)
+        BOOST_ASSERT(impl_->synchronizing_stage[source] >= stage);
+
+      // receive any batches sent in the meantime
+      // all have to be available already
+      while (true) {
+        bool done=true;
+        for (int source=0; source<p ; ++source)
+          if(impl_->number_received_batches[source] < 0)
+            done = false;
+        if (done)
+          break;
+        poll(false,-1,true);
+      }
+      
+#ifndef NO_IMMEDIATE_PROCESSING
+      for (int source=0; source<p ; ++source)
+        BOOST_ASSERT(impl_->number_received_batches[source] >= 0);
+#endif
+
+      impl_->synchronizing = false;
+      
+      // Flush out remaining messages
+      if (impl_->synchronizing_unfinished[stage]==0)
+        break;
+#ifdef NO_IMMEDIATE_PROCESSING
+      for (process_id_type dest = 0; dest < p; ++dest)
+        process_batch(dest);
+#endif
+
+      no_new_messages = true;
+      for (process_id_type dest = 0; dest < p; ++dest) {
+        if (impl_->outgoing[dest].headers.size() || 
+            impl_->number_sent_batches[dest]>0)
+          no_new_messages = false;
+        send_batch(dest);
+      }
+    }
+
+  impl_->comm.barrier/*nomacro*/();
+#if 0
+  // set up for next synchronize call
+  for (int source=0; source<p; ++source) {
+    if (impl_->synchronizing_stage[source] != stage) {
+      std::cerr << id << ": expecting stage " << stage << " from source "
+                << source << ", got " << impl_->synchronizing_stage[source]
+                << std::endl;
+    }
+    BOOST_ASSERT(impl_->synchronizing_stage[source]==stage);
+  }
+#endif
+  std::fill(impl_->synchronizing_stage.begin(),
+            impl_->synchronizing_stage.end(), -1);
+            
+  // get rid of the information regarding recorded numbers of processors
+  // for the stages we just finished
+  impl_->processors_synchronizing_stage.clear();
+  impl_->synchronizing_unfinished.clear();
+
+  for (process_id_type dest = 0; dest < p; ++dest)
+    BOOST_ASSERT (impl_->outgoing[dest].headers.empty());
+#ifndef NO_IMMEDIATE_PROCESSING
+      for (int source=0; source<p ; ++source)
+        BOOST_ASSERT (impl_->number_received_batches[source] == 0);
+#endif
+
+  impl_->free_sent_batches();
+#ifdef DEBUG
+  std::cerr << "SYNC: " << process_id(*this) << " completed." << std::endl;
+#endif
+}
+
+optional<std::pair<mpi_process_group::process_id_type, int> >
+probe(const mpi_process_group& pg)
+{ return pg.probe(); }
+
+void mpi_process_group::poll_requests(int block) const
+{
+  int size = impl_->requests.size();
+  if (size==0)
+    return;
+  std::vector<MPI_Status> statuses(size);
+  std::vector<int> indices(size);
+  
+  while (true) {
+    MPI_Testsome(impl_->requests.size(),&impl_->requests[0],
+       &size,&indices[0],&statuses[0]);
+    if (size==0)
+      return; // no message waiting
+
+    // remove handled requests before we get the chance to be recursively called
+    if (size) {
+      std::vector<MPI_Request> active_requests;
+      std::size_t i=0;
+      int j=0;
+      for (;i< impl_->requests.size() && j< size; ++i) {
+        if (int(i)==indices[j])
+          // release the dealt-with request 
+          ++j;
+        else // copy and keep the request
+          active_requests.push_back(impl_->requests[i]);
+      }    
+      while (i < impl_->requests.size())
+        active_requests.push_back(impl_->requests[i++]);
+      impl_->requests.swap(active_requests);
+    }
+
+    optional<std::pair<int, int> > result;
+    for (int i=0;i < size; ++i) {
+      std::pair<int, int> decoded = decode_tag(statuses[i].MPI_TAG);
+      block_type* block = impl_->blocks[decoded.first];
+      
+      BOOST_ASSERT (decoded.second < static_cast<int>(block->triggers.size()) && block->triggers[decoded.second]);
+        // We have a trigger for this message; use it
+      trigger_receive_context old_context = impl_->trigger_context;
+      impl_->trigger_context = trc_irecv_out_of_band;
+      block->triggers[decoded.second]->receive(*this, statuses[i].MPI_SOURCE, 
+            decoded.second, impl_->trigger_context, decoded.first);
+      impl_->trigger_context = old_context;
+    }
+  }
+}
+
+
+optional<std::pair<int, int> > 
+mpi_process_group::
+poll(bool wait, int block, bool synchronizing) const
+{
+  // Set the new trigger context for these receive operations
+  trigger_receive_context old_context = impl_->trigger_context;
+  if (synchronizing)
+    impl_->trigger_context = trc_in_synchronization;
+  else
+    impl_->trigger_context = trc_out_of_band;
+
+  //wait = false;
+  optional<boost::mpi::status> status;
+  bool finished=false;
+  optional<std::pair<int, int> > result;
+  do {
+    poll_requests(block);
+    // Check for any messages not yet received.
+#ifdef PBGL_PROCESS_GROUP_NO_IRECV
+    if (wait)
+      status = impl_->comm.probe();
+    else 
+#endif
+       status = impl_->comm.iprobe();
+
+    if (status) { // we have a message
+      // Decode the message
+      std::pair<int, int> decoded = decode_tag(status.get().tag());
+      if (maybe_emit_receive(status.get().source(), status.get().tag()))
+        // We received the message out-of-band; poll again
+        finished = false;
+      else if (decoded.first == (block == -1 ? my_block_number() : block)) {
+        // This message is for us, but not through a trigger. Return
+        // the decoded message.
+        result = std::make_pair(status.get().source(), decoded.second);
+      // otherwise we didn't match any message we know how to deal with, so
+      // pretend no message exists.
+        finished = true;
+      }
+    }
+    else
+      // We don't have a message to process; we're done.
+      finished=!wait;
+  } while (!finished);
+
+  // Restore the prior trigger context
+  impl_->trigger_context = old_context;
+  poll_requests(block);
+  return result;
+}
+
+void synchronize(const mpi_process_group& pg) { pg.synchronize(); }
+
+mpi_process_group mpi_process_group::base() const
+{
+  mpi_process_group copy(*this);
+  copy.block_num.reset();
+  return copy;
+}
+  
+
+void mpi_process_group::impl::free_sent_batches()
+{
+  typedef std::list<batch_request>::iterator iterator;
+  iterator it = sent_batches.begin();
+  int flag;
+  int result;
+  while(it != sent_batches.end()) {
+    result = MPI_Test(&it->request,&flag,MPI_STATUS_IGNORE);
+    BOOST_ASSERT(result == MPI_SUCCESS);
+    iterator next=it;
+    ++next;
+    if (flag)
+      sent_batches.erase(it);
+    it=next;
+  }  
+#ifdef PREALLOCATE_BATCHES
+  for (std::size_t i=0; i< batch_pool.size();++i) {
+    if(batch_pool[i].request != MPI_REQUEST_NULL) {
+      result = MPI_Test(&batch_pool[i].request,&flag,MPI_STATUS_IGNORE);
+      BOOST_ASSERT(result == MPI_SUCCESS);
+      if (flag) {
+        free_batches.push(i);
+        batch_pool[i].request = MPI_REQUEST_NULL;
+        batch_pool[i].buffer.resize(0);
+      }
+    }
+  } 
+#endif
+}
+
+void 
+mpi_process_group::install_trigger(int tag, int block, 
+      shared_ptr<trigger_base> const& launcher)
+{
+  block_type* my_block = impl_->blocks[block];
+  BOOST_ASSERT(my_block);
+
+  // Make sure we have enough space in the structure for this trigger.
+  if (launcher->tag() >= static_cast<int>(my_block->triggers.size()))
+    my_block->triggers.resize(launcher->tag() + 1);
+
+  // If someone already put a trigger in this spot, we have a big
+  // problem.
+  if (my_block->triggers[launcher->tag()]) {
+    std::cerr << "Block " << my_block_number() 
+              << " already has a trigger for tag " << launcher->tag()
+              << std::endl;
+  }
+  BOOST_ASSERT(!my_block->triggers[launcher->tag()]);
+
+  // Attach a new trigger launcher
+  my_block->triggers[launcher->tag()] = launcher;
+}
+
+std::size_t mpi_process_group::message_buffer_size()
+{
+  return message_buffer.size();
+}
+
+
+void mpi_process_group::set_message_buffer_size(std::size_t s)
+{
+  int sz;
+  void* ptr;
+  if (!message_buffer.empty()) {
+    MPI_Buffer_detach(&ptr,&sz);
+    BOOST_ASSERT(ptr == &message_buffer.front());
+    BOOST_ASSERT(static_cast<std::size_t>(sz)  == message_buffer.size());
+  }
+  else if (old_buffer != 0)
+    MPI_Buffer_detach(&old_buffer,&old_buffer_size);
+  message_buffer.resize(s);
+  if (s)
+    MPI_Buffer_attach(&message_buffer.front(), message_buffer.size());
+  else if (old_buffer_size)
+    MPI_Buffer_attach(old_buffer, old_buffer_size);
+}
+
+
+std::vector<char> mpi_process_group::message_buffer;
+int mpi_process_group::old_buffer_size=0;
+void* mpi_process_group::old_buffer=0;
+
+} } } // end namespace boost::graph::distributed
diff --git a/src/boost/libs/graph_parallel/src/tag_allocator.cpp b/src/boost/libs/graph_parallel/src/tag_allocator.cpp
new file mode 100644
index 00000000..a89a2a2d
--- /dev/null
+++ b/src/boost/libs/graph_parallel/src/tag_allocator.cpp
@@ -0,0 +1,45 @@
+// Copyright (C) 2007  Douglas Gregor  <doug.gregor@gmail.com>
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+#include <boost/graph/use_mpi.hpp>
+#include <boost/graph/distributed/detail/tag_allocator.hpp>
+
+namespace boost { namespace graph { namespace distributed { namespace detail {
+
+tag_allocator::token tag_allocator::get_tag() 
+{
+  int tag;
+
+  if (!freed.empty()) {
+    // Grab the tag at the top of the stack.
+    tag = freed.back();
+    freed.pop_back();
+  } else
+    // Grab the bottom tag and move the bottom down
+    tag = bottom--;
+
+  return token(this, tag);
+}
+
+tag_allocator::token::token(const token& other)
+  : allocator(other.allocator), tag_(other.tag_)
+{
+  // other no longer owns this tag
+  other.tag_ = -1;
+}
+
+tag_allocator::token::~token()
+{
+  if (tag_ != -1) {
+    if (tag_ == allocator->bottom + 1)
+      // This is the bottommost tag: just bump up the bottom
+      ++allocator->bottom;
+    else
+      // This tag isn't the bottom, so push it only the freed stack
+      allocator->freed.push_back(tag_);
+  }
+}
+
+} } } } // end namespace boost::graph::distributed::detail
diff --git a/src/boost/libs/graph_parallel/test/Jamfile.v2 b/src/boost/libs/graph_parallel/test/Jamfile.v2
new file mode 100644
index 00000000..61bf1618
--- /dev/null
+++ b/src/boost/libs/graph_parallel/test/Jamfile.v2
@@ -0,0 +1,47 @@
+#
+# (C) Copyright 2005, 2006 Trustees of Indiana University
+# (C) Copyright 2005 Douglas Gregor
+#
+# Distributed under the Boost Software License, Version 1.0. (See accompanying 
+# file LICENSE_1_0.txt or copy at http://www.boost.org/LICENSE_1_0.txt.)
+
+
+# use-project /boost/mpi : ../build ;
+
+project /boost/graph_parallel/test 
+    : requirements <library>../build//boost_graph_parallel <library>../../system/build//boost_system ;
+    
+import mpi : mpi-test ;
+
+if [ mpi.configured ]
+{
+test-suite graph_parallel
+  :
+  [ mpi-test distributed_property_map_test : : : 2 ]
+  [ mpi-test distributed_queue_test : : : 2 ]
+  [ mpi-test process_group_serialization : : : 2 ]
+  [ mpi-test adjlist_build_test : : : 2 ]
+  [ mpi-test adjlist_redist_test : : : 2 ]
+  [ mpi-test adjlist_remove_test : : : 2 ]
+  [ mpi-test distributed_adjacency_list_test : : : 2 ]
+  [ mpi-test distributed_connected_components_test : : : 2 ]
+  [ mpi-test distributed_page_rank_test : : : 2 ]
+  [ mpi-test distributed_csr_test : : : 2 ]
+  [ mpi-test distributed_dfs_test : : : 2 ]
+  [ mpi-test distributed_graph_coloring_test : : : 2 ]
+  [ mpi-test distributed_mst_test : : : 2 ]
+  [ mpi-test distributed_strong_components_test : : : 2 ]
+  [ mpi-test hohberg_biconnected_components_test : : : 2 ]
+  [ mpi-test mesh_generator_test : : <testing.arg>"1000 1000 1 0" : 2 ]
+  [ mpi-test named_vertices_seq : : : 1 ]
+  [ mpi-test distributed_shortest_paths_test : : : 2 ]
+  [ mpi-test distributed_csr_algorithm_test : : : 1 ]
+  [ mpi-test distributed_betweenness_centrality_test : : : 2 ]
+  [ mpi-test distributed_dimacs_reader : : : 2 ]
+  [ mpi-test distributed_rmat_cc_ps : : : 2 ]
+  [ mpi-test distributed_rmat_cc : : : 2 ]
+  [ mpi-test distributed_rmat_pagerank : : : 2 ]
+  [ mpi-test distributed_st_connected_test : : : 2 ]
+  ;
+}
+
diff --git a/src/boost/libs/graph_parallel/test/adjlist_build_test.cpp b/src/boost/libs/graph_parallel/test/adjlist_build_test.cpp
new file mode 100644
index 00000000..87335a50
--- /dev/null
+++ b/src/boost/libs/graph_parallel/test/adjlist_build_test.cpp
@@ -0,0 +1,245 @@
+// Copyright (C) 2007  Douglas Gregor
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+#include <boost/graph/use_mpi.hpp>
+#include <boost/config.hpp>
+#include <boost/throw_exception.hpp>
+#include <boost/graph/distributed/adjacency_list.hpp>
+#include <boost/graph/distributed/mpi_process_group.hpp>
+#include <boost/test/minimal.hpp>
+#include <boost/random/linear_congruential.hpp>
+#include <boost/graph/erdos_renyi_generator.hpp>
+#include <boost/lexical_cast.hpp>
+#include <ctime>
+
+using namespace boost;
+using boost::graph::distributed::mpi_process_group;
+
+#ifdef BOOST_NO_EXCEPTIONS
+void
+boost::throw_exception(std::exception const& ex)
+{
+    std::cout << ex.what() << std::endl;
+    abort();
+}
+#endif
+
+
+int test_main(int argc, char** argv)
+{
+  boost::mpi::environment env(argc, argv);
+
+  int n = 10000;
+  double p = 3e-3;
+  int seed = std::time(0);
+  int immediate_response_percent = 10;
+
+  if (argc > 1) n = lexical_cast<int>(argv[1]);
+  if (argc > 2) p = lexical_cast<double>(argv[2]);
+  if (argc > 3) seed = lexical_cast<int>(argv[3]);
+
+  typedef adjacency_list<listS, 
+                         distributedS<mpi_process_group, vecS>,
+                         bidirectionalS> Graph;
+
+  mpi_process_group pg;
+  int rank = process_id(pg);
+  int numprocs = num_processes(pg);
+  bool i_am_root = rank == 0;
+
+  // broadcast the seed  
+  broadcast(pg, seed, 0);
+
+  // Random number generator
+  minstd_rand gen;
+  minstd_rand require_response_gen;
+
+  if (i_am_root) {
+    std::cout << "n = " << n << ", p = " << p << ", seed = " << seed
+              << "\nBuilding graph with the iterator constructor... ";
+    std::cout.flush();
+  }
+
+  // Build a graph using the iterator constructor, where each of the
+  // processors has exactly the same information.
+  gen.seed(seed);
+  Graph g1(erdos_renyi_iterator<minstd_rand, Graph>(gen, n, p),
+           erdos_renyi_iterator<minstd_rand, Graph>(),
+           n, pg, Graph::graph_property_type());
+  // NGE: Grrr, the default graph property is needed to resolve an 
+  //      ambiguous overload in the adjaceny list constructor
+
+  // Build another, identical graph using add_edge
+  if (i_am_root) {
+    std::cout << "done.\nBuilding graph with add_edge from the root...";
+    std::cout.flush();
+  }
+
+  gen.seed(seed);
+  require_response_gen.seed(1);
+  Graph g2(n, pg);
+  if (i_am_root) {
+    // The root will add all of the edges, some percentage of which
+    // will require an immediate response from the owner of the edge.
+    for (erdos_renyi_iterator<minstd_rand, Graph> first(gen, n, p), last;
+         first != last; ++first) {
+      Graph::lazy_add_edge lazy
+        = add_edge(vertex(first->first, g2), vertex(first->second, g2), g2);
+
+      if ((int)require_response_gen() % 100 < immediate_response_percent) {
+        // Send out-of-band to require a response 
+        std::pair<graph_traits<Graph>::edge_descriptor, bool> result(lazy);
+        BOOST_CHECK(source(result.first, g2) == vertex(first->first, g2));
+        BOOST_CHECK(target(result.first, g2) == vertex(first->second, g2));
+      }
+    }
+  }
+
+  if (i_am_root) {
+    std::cout << "synchronizing...";
+    std::cout.flush();
+  }
+
+  synchronize(g2);
+
+  // Verify that the two graphs are indeed identical.
+  if (i_am_root) {
+    std::cout << "done.\nVerifying graphs...";
+    std::cout.flush();
+  }
+
+  // Check the number of vertices
+  if (num_vertices(g1) != num_vertices(g2)) {
+    std::cerr << g1.processor() << ": g1 has " << num_vertices(g1) 
+              << " vertices, g2 has " << num_vertices(g2) << " vertices.\n";
+    communicator(pg).abort(-1);
+  }
+
+  // Check the number of edges
+  if (num_edges(g1) != num_edges(g2)) {
+    std::cerr << g1.processor() << ": g1 has " << num_edges(g1) 
+              << " edges, g2 has " << num_edges(g2) << " edges.\n";
+    communicator(pg).abort(-1);
+  }
+
+  // Check the in-degree and out-degree of each vertex
+  graph_traits<Graph>::vertex_iterator vfirst1, vlast1, vfirst2, vlast2;
+  boost::tie(vfirst1, vlast1) = vertices(g1);
+  boost::tie(vfirst2, vlast2) = vertices(g2);
+  for(; vfirst1 != vlast1 && vfirst2 != vlast2; ++vfirst1, ++vfirst2) {
+    if (out_degree(*vfirst1, g1) != out_degree(*vfirst2, g2)) {
+      std::cerr << g1.processor() << ": out-degree mismatch ("
+                << out_degree(*vfirst1, g1) << " vs. " 
+                << out_degree(*vfirst2, g2) << ").\n";
+      communicator(pg).abort(-1);
+    }
+
+    if (in_degree(*vfirst1, g1) != in_degree(*vfirst2, g2)) {
+      std::cerr << g1.processor() << ": in-degree mismatch (" 
+                << in_degree(*vfirst1, g1) << " vs. " 
+                << in_degree(*vfirst2, g2) << ").\n";
+      communicator(pg).abort(-1);
+    }
+  }
+
+  // TODO: Check the actual edge targets
+
+  // Build another, identical graph using add_edge
+  if (i_am_root) {
+    std::cout << "done.\nBuilding graph with add_edge from everywhere...";
+    std::cout.flush();
+  }
+
+  gen.seed(seed);
+  require_response_gen.seed(1);
+  Graph g3(n, pg);
+
+  {
+    // Each processor will take a chunk of incoming edges and add
+    // them. Some percentage of the edge additions will require an
+    // immediate response from the owner of the edge. This should
+    // create a lot of traffic when building the graph, but should
+    // produce a graph identical to the other graphs.
+    typedef graph_traits<Graph>::edges_size_type edges_size_type;
+
+    erdos_renyi_iterator<minstd_rand, Graph> first(gen, n, p);
+    edges_size_type chunk_size = edges_size_type(p*n*n)/numprocs;
+    edges_size_type start = chunk_size * rank;
+    edges_size_type remaining_edges = 
+      (rank < numprocs - 1? chunk_size 
+       : edges_size_type(p*n*n) - start);
+
+    // Spin the generator to the first edge we're responsible for
+    for (; start; ++first, --start) ;
+
+    for (; remaining_edges; --remaining_edges, ++first) {
+      Graph::lazy_add_edge lazy
+        = add_edge(vertex(first->first, g3), vertex(first->second, g3), g3);
+
+      if ((int)require_response_gen() % 100 < immediate_response_percent) {
+        // Send out-of-band to require a response 
+        std::pair<graph_traits<Graph>::edge_descriptor, bool> result(lazy);
+        BOOST_CHECK(source(result.first, g3) == vertex(first->first, g3));
+        BOOST_CHECK(target(result.first, g3) == vertex(first->second, g3));
+      }
+    }
+  }
+
+  if (i_am_root) {
+    std::cout << "synchronizing...";
+    std::cout.flush();
+  }
+
+  synchronize(g3);
+
+  // Verify that the two graphs are indeed identical.
+  if (i_am_root) {
+    std::cout << "done.\nVerifying graphs...";
+    std::cout.flush();
+  }
+
+  // Check the number of vertices
+  if (num_vertices(g1) != num_vertices(g3)) {
+    std::cerr << g1.processor() << ": g1 has " << num_vertices(g1) 
+              << " vertices, g3 has " << num_vertices(g3) << " vertices.\n";
+    communicator(pg).abort(-1);
+  }
+
+  // Check the number of edges
+  if (num_edges(g1) != num_edges(g3)) {
+    std::cerr << g1.processor() << ": g1 has " << num_edges(g1) 
+              << " edges, g3 has " << num_edges(g3) << " edges.\n";
+    communicator(pg).abort(-1);
+  }
+
+  // Check the in-degree and out-degree of each vertex
+  boost::tie(vfirst1, vlast1) = vertices(g1);
+  boost::tie(vfirst2, vlast2) = vertices(g3);
+  for(; vfirst1 != vlast1 && vfirst2 != vlast2; ++vfirst1, ++vfirst2) {
+    if (out_degree(*vfirst1, g1) != out_degree(*vfirst2, g3)) {
+      std::cerr << g1.processor() << ": out-degree mismatch ("
+                << out_degree(*vfirst1, g1) << " vs. " 
+                << out_degree(*vfirst2, g3) << ").\n";
+      communicator(pg).abort(-1);
+    }
+
+    if (in_degree(*vfirst1, g1) != in_degree(*vfirst2, g3)) {
+      std::cerr << g1.processor() << ": in-degree mismatch (" 
+                << in_degree(*vfirst1, g1) << " vs. " 
+                << in_degree(*vfirst2, g3) << ").\n";
+      communicator(pg).abort(-1);
+    }
+  }
+
+  // TODO: Check the actual edge targets
+
+  if (i_am_root) {
+    std::cout << "done.\n";
+    std::cout.flush();
+  }
+
+  return 0;
+}
diff --git a/src/boost/libs/graph_parallel/test/adjlist_redist_test.cpp b/src/boost/libs/graph_parallel/test/adjlist_redist_test.cpp
new file mode 100644
index 00000000..96a659ae
--- /dev/null
+++ b/src/boost/libs/graph_parallel/test/adjlist_redist_test.cpp
@@ -0,0 +1,216 @@
+// Copyright (C) 2005, 2006 The Trustees of Indiana University.
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+//  Authors: Douglas Gregor
+//           Andrew Lumsdaine
+
+#include <boost/graph/use_mpi.hpp>
+#include <boost/config.hpp>
+#include <boost/throw_exception.hpp>
+#include <boost/lexical_cast.hpp>
+#include <boost/serialization/vector.hpp>
+#include <boost/graph/distributed/adjacency_list.hpp>
+#include <boost/graph/distributed/mpi_process_group.hpp>
+#include <boost/random/linear_congruential.hpp>
+#include <iostream>
+#include <boost/property_map/property_map_iterator.hpp>
+#include <boost/graph/distributed/dehne_gotz_min_spanning_tree.hpp>
+#include <boost/graph/distributed/vertex_list_adaptor.hpp>
+#include <boost/graph/erdos_renyi_generator.hpp>
+#include <boost/graph/distributed/graphviz.hpp>
+#include <sstream>
+#include <string>
+#include <boost/graph/iteration_macros.hpp>
+#include <boost/test/minimal.hpp>
+
+#ifdef BOOST_NO_EXCEPTIONS
+void
+boost::throw_exception(std::exception const& ex)
+{
+    std::cout << ex.what() << std::endl;
+    abort();
+}
+#endif
+
+using namespace boost;
+using boost::graph::distributed::mpi_process_group;
+
+/****************************************************************************
+ * Edge weight generator iterator                                           *
+ ****************************************************************************/
+template<typename F>
+class generator_iterator
+{
+public:
+  typedef std::input_iterator_tag iterator_category;
+  typedef typename F::result_type value_type;
+  typedef const value_type&       reference;
+  typedef const value_type*       pointer;
+  typedef void                    difference_type;
+
+  explicit generator_iterator(const F& f = F()) : f(f) { value = this->f(); }
+
+  reference operator*() const  { return value; }
+  pointer   operator->() const { return &value; }
+
+  generator_iterator& operator++()
+  {
+    value = f();
+    return *this;
+  }
+
+  generator_iterator operator++(int)
+  {
+    generator_iterator temp(*this);
+    ++(*this);
+    return temp;
+  }
+
+  bool operator==(const generator_iterator& other) const
+  { return f == other.f; }
+
+  bool operator!=(const generator_iterator& other) const
+  { return !(*this == other); }
+
+private:
+  F f;
+  value_type value;
+};
+
+template<typename F>
+inline generator_iterator<F> make_generator_iterator(const F& f)
+{ return generator_iterator<F>(f); }
+
+typedef minstd_rand RandomGenerator;
+
+template<typename Graph>
+double get_mst_weight (const Graph& g)
+{
+  typedef typename graph_traits<Graph>::edge_descriptor edge_descriptor;
+  typename property_map<Graph, edge_weight_t>::const_type weight 
+    = get(edge_weight, g);
+
+  // Boruvka then merge test
+  std::vector<edge_descriptor> results;
+  graph::boruvka_then_merge(make_vertex_list_adaptor(g), weight,
+                            std::back_inserter(results));
+  if (process_id(g.process_group()) == 0)
+    return accumulate(make_property_map_iterator(weight, results.begin()),
+                      make_property_map_iterator(weight, results.end()),
+                      0.0);
+  else
+    return 0.0;
+}
+
+template<typename Graph>
+void test_redistribution(int n, double p, int iterations, bool debug_output)
+{
+  RandomGenerator gen;
+  Graph g(erdos_renyi_iterator<RandomGenerator, Graph>(gen, n, p),
+          erdos_renyi_iterator<RandomGenerator, Graph>(),
+          make_generator_iterator(uniform_01<RandomGenerator>(gen)),
+          n);
+
+  int iter = 0;
+  mpi_process_group pg = g.process_group();
+
+  // Set the names of the vertices to be the global index in the
+  // initial distribution. Then when we are debugging we'll be able to
+  // see how vertices have moved.
+  {
+    typedef typename property_map<Graph, vertex_index_t>::type VertexIndexMap;
+    typedef typename property_map<Graph, vertex_global_t>::type VertexGlobalMap;
+    typename property_map<Graph, vertex_name_t>::type name_map 
+      = get(vertex_name, g);
+
+    parallel::global_index_map<VertexIndexMap, VertexGlobalMap> 
+      global_index(g.process_group(), num_vertices(g), 
+                   get(vertex_index, g), get(vertex_global, g));
+    BGL_FORALL_VERTICES_T(v, g, Graph)
+      put(name_map, v, get(global_index, v));
+  }
+
+  if (debug_output)
+    write_graphviz("redist-0.dot", g,
+                   make_label_writer(get(vertex_name, g)),
+                   make_label_writer(get(edge_weight, g)));
+
+  double mst_weight = get_mst_weight(g);
+  if (process_id(pg) == 0)
+    std::cout << "MST weight = " << mst_weight << std::endl;
+
+  RandomGenerator nonsync_gen(process_id(pg) + gen());
+  while (++iter <= iterations) {
+    typename property_map<Graph, vertex_rank_t>::type to_processor_map = 
+      get(vertex_rank, g);
+
+    // Randomly assign a new distribution
+    typename graph_traits<Graph>::vertex_iterator vi, vi_end;
+    for (boost::tie(vi, vi_end) = vertices(g); vi != vi_end; ++vi)
+      put(to_processor_map, *vi, gen() % num_processes(pg));
+
+    if (process_id(pg) == 0)
+      std::cout << "Redistributing...";
+    // Perform the actual redistribution
+    g.redistribute(to_processor_map);
+
+    if (process_id(pg) == 0)
+      std::cout << " done." << std::endl;
+
+    if (debug_output) {
+      std::ostringstream out;
+      out << "redist-" << iter << ".dot";
+      write_graphviz(out.str().c_str(), g,
+                     make_label_writer(get(vertex_name, g)),
+                     make_label_writer(get(edge_weight, g)));
+    }
+
+    // Check that the MST weight is unchanged
+    double new_mst_weight = get_mst_weight(g);
+    if (process_id(pg) == 0) {
+      std::cout << "MST weight = " << new_mst_weight << std::endl;
+      if (std::fabs(new_mst_weight - mst_weight) > 0.0001)
+        communicator(pg).abort(-1);    }
+  }
+}
+
+int test_main(int argc, char** argv)
+{
+  int n = 1000;
+  double p = 3e-3;
+  int iterations = 5;
+  bool debug_output = false;
+
+  boost::mpi::environment env(argc, argv);
+
+  if (argc > 1) n = lexical_cast<int>(argv[1]);
+  if (argc > 2) p = lexical_cast<double>(argv[2]);
+  if (argc > 3) iterations = lexical_cast<int>(argv[3]);
+  if (argc > 4) debug_output = true;
+
+  typedef adjacency_list<listS, 
+                         distributedS<mpi_process_group, vecS>,
+                         undirectedS,
+                         // Vertex properties
+                         property<vertex_name_t, std::size_t,
+                           property<vertex_rank_t, int> >,
+                         // Edge properties
+                         property<edge_weight_t, double> > UnstableUDGraph;
+  typedef adjacency_list<listS, 
+                         distributedS<mpi_process_group, listS>,
+                         undirectedS,
+                         // Vertex properties
+                         property<vertex_name_t, std::size_t,
+                           property<vertex_rank_t, int,
+                              property<vertex_index_t, std::size_t> > >,
+                         // Edge properties
+                         property<edge_weight_t, double> > StableUDGraph;
+
+  test_redistribution<UnstableUDGraph>(n, p, iterations, debug_output);
+  test_redistribution<StableUDGraph>(n, p, iterations, debug_output);
+
+  return 0;
+}
diff --git a/src/boost/libs/graph_parallel/test/adjlist_remove_test.cpp b/src/boost/libs/graph_parallel/test/adjlist_remove_test.cpp
new file mode 100644
index 00000000..549c5195
--- /dev/null
+++ b/src/boost/libs/graph_parallel/test/adjlist_remove_test.cpp
@@ -0,0 +1,146 @@
+// Copyright 2004 The Trustees of Indiana University.
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+//  Authors: Douglas Gregor
+//           Andrew Lumsdaine
+
+#include <boost/graph/use_mpi.hpp>
+#include <boost/config.hpp>
+#include <boost/throw_exception.hpp>
+#include <boost/graph/distributed/adjacency_list.hpp>
+#include <boost/graph/distributed/mpi_process_group.hpp>
+#include <boost/graph/parallel/distribution.hpp>
+#include <iostream>
+#include <cassert>
+#include <boost/test/minimal.hpp>
+
+#ifdef BOOST_NO_EXCEPTIONS
+void
+boost::throw_exception(std::exception const& ex)
+{
+    std::cout << ex.what() << std::endl;
+    abort();
+}
+#endif
+
+using namespace boost;
+using boost::graph::distributed::mpi_process_group;
+
+void test_bidirectional_graph()
+{
+  typedef adjacency_list<listS, distributedS<mpi_process_group, vecS>,
+                         bidirectionalS> Graph;
+  typedef graph_traits<Graph>::vertex_descriptor Vertex;
+
+  typedef std::pair<std::size_t, std::size_t> E;
+  E edges[3] = { E(0,3), E(1,4), E(2,5) };
+  
+  Graph g(&edges[0], &edges[0] + 3, 6);
+  assert(num_processes(g.process_group()) == 2);
+
+  if (process_id(g.process_group()) == 0)
+    std::cerr << "Testing bidirectional graph edge removal...";
+  synchronize(g.process_group());
+
+  graph_traits<Graph>::vertex_iterator vi = vertices(g).first;
+  Vertex u = *vi++;
+  Vertex v = *vi++;
+  Vertex w = *vi++;
+  BOOST_CHECK(vi == vertices(g).second);
+  
+  BOOST_CHECK((process_id(g.process_group()) == 0 && out_degree(u, g) == 1)
+             || (process_id(g.process_group()) == 1 && in_degree(u, g) == 1));
+  
+  BOOST_CHECK((process_id(g.process_group()) == 0 && out_degree(v, g) == 1)
+             || (process_id(g.process_group()) == 1 && in_degree(v, g) == 1));
+
+  BOOST_CHECK((process_id(g.process_group()) == 0 && out_degree(w, g) == 1)
+             || (process_id(g.process_group()) == 1 && in_degree(w, g) == 1));
+
+  // Remove edges
+  if (process_id(g.process_group()) == 0) {
+    remove_edge(*out_edges(u, g).first, g);
+    remove_edge(*out_edges(w, g).first, g);
+  } else {
+    remove_edge(*in_edges(v, g).first, g);
+    remove_edge(*in_edges(w, g).first, g);
+  }
+  
+  synchronize(g);
+
+  BOOST_CHECK(num_edges(g) == 0);
+  BOOST_CHECK(out_degree(u, g) == 0);
+  BOOST_CHECK(out_degree(v, g) == 0);
+  BOOST_CHECK(out_degree(w, g) == 0);
+  BOOST_CHECK(in_degree(u, g) == 0);
+  BOOST_CHECK(in_degree(v, g) == 0);
+  BOOST_CHECK(in_degree(w, g) == 0);
+
+  if (process_id(g.process_group()) == 0) std::cerr << "OK.\n";
+}
+
+void test_undirected_graph()
+{
+  typedef adjacency_list<listS, distributedS<mpi_process_group, vecS>,
+                         undirectedS> Graph;
+  typedef graph_traits<Graph>::vertex_descriptor Vertex;
+  typedef graph_traits<Graph>::edge_descriptor Edge;
+
+  typedef std::pair<std::size_t, std::size_t> E;
+  E the_edges[3] = { E(0,3), E(1,4), E(2,5) };
+  
+  Graph g(&the_edges[0], &the_edges[0] + 3, 6);
+  assert(num_processes(g.process_group()) == 2);
+
+  if (process_id(g.process_group()) == 0)
+    std::cerr << "Testing undirected graph edge removal...";
+  synchronize(g.process_group());
+
+  graph_traits<Graph>::vertex_iterator vi = vertices(g).first;
+  Vertex u = *vi++;
+  Vertex v = *vi++;
+  Vertex w = *vi++;
+  BOOST_CHECK(vi == vertices(g).second);  
+  BOOST_CHECK(degree(u, g) == 1);
+  BOOST_CHECK(degree(v, g) == 1);
+  BOOST_CHECK(degree(w, g) == 1);
+  
+  // Remove edges
+  if (process_id(g.process_group()) == 0) {
+    BOOST_CHECK(num_edges(g) == 3);
+    remove_edge(*out_edges(u, g).first, g);
+    remove_edge(*out_edges(w, g).first, g);
+  } else {
+    BOOST_CHECK(num_edges(g) == 0);
+    remove_edge(*in_edges(v, g).first, g);
+    remove_edge(*in_edges(w, g).first, g);
+  }
+  
+  synchronize(g);
+
+  if (num_edges(g) > 0) {
+    Edge e = *edges(g).first;
+    std::cerr << "#" << process_id(g.process_group()) << ": extra edge! ("
+              << local(source(e, g)) << "@" << owner(source(e, g))
+              << ", " 
+              << local(target(e, g)) << "@" << owner(target(e, g))
+              << ")\n";
+  }
+
+  BOOST_CHECK(num_edges(g) == 0);
+  BOOST_CHECK(degree(u, g) == 0);
+  BOOST_CHECK(degree(v, g) == 0);
+  BOOST_CHECK(degree(w, g) == 0);
+  if (process_id(g.process_group()) == 0) std::cerr << "OK.\n";
+}
+
+int test_main(int argc, char** argv)
+{
+  boost::mpi::environment env(argc, argv);
+  test_bidirectional_graph();
+  test_undirected_graph();
+  return 0;
+}
diff --git a/src/boost/libs/graph_parallel/test/algorithm_performance.cpp b/src/boost/libs/graph_parallel/test/algorithm_performance.cpp
new file mode 100644
index 00000000..c4973b41
--- /dev/null
+++ b/src/boost/libs/graph_parallel/test/algorithm_performance.cpp
@@ -0,0 +1,887 @@
+// Copyright 2004 The Trustees of Indiana University.
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+//  Authors: Nick Edmonds
+//           Andrew Lumsdaine
+
+// #define PBGL_ACCOUNTING
+
+#include <boost/graph/use_mpi.hpp>
+
+#include <boost/graph/distributed/compressed_sparse_row_graph.hpp>
+#include <boost/graph/distributed/adjacency_list.hpp>
+
+#include <boost/graph/distributed/mpi_process_group.hpp>
+
+#include <boost/test/minimal.hpp>
+#include <boost/random.hpp>
+#include <boost/property_map/parallel/distributed_property_map.hpp>
+#include <boost/graph/distributed/graphviz.hpp>
+#include <boost/graph/iteration_macros.hpp>
+#include <boost/graph/properties.hpp>
+
+#include <boost/graph/rmat_graph_generator.hpp>
+#include <boost/graph/small_world_generator.hpp>
+#include <boost/graph/erdos_renyi_generator.hpp>
+
+#include <boost/graph/distributed/connected_components.hpp>
+#include <boost/graph/distributed/connected_components_parallel_search.hpp>
+#include <boost/graph/distributed/betweenness_centrality.hpp>
+#include <boost/graph/distributed/delta_stepping_shortest_paths.hpp>
+
+#include <time.h>
+#include <sys/time.h>
+
+#include <iostream>
+#include <iomanip>
+#include <vector>
+#include <stdint.h>
+
+// Edge distribution tags select a generator
+struct rmat_edge_distribution_tag { };
+struct rmat_unique_edge_distribution_tag { };
+
+using namespace boost;
+using boost::graph::distributed::mpi_process_group;
+
+/****************************************************************************
+ * Timing
+ ****************************************************************************/
+#ifndef PBGL_ACCOUNTING 
+
+typedef double time_type;
+
+inline time_type get_time()
+{
+  timeval tp;
+  gettimeofday(&tp, 0);
+  return tp.tv_sec + tp.tv_usec / 1000000.0;
+}
+
+std::string print_time(time_type t)
+{
+  std::ostringstream out;
+  out << std::setiosflags(std::ios::fixed) << std::setprecision(2) << t;
+  return out.str();
+}
+
+#endif // PBGL_ACCOUNTING
+
+/****************************************************************************
+ * Edge weight generator iterator                                           *
+ ****************************************************************************/
+template<typename F, typename RandomGenerator>
+class generator_iterator
+{
+public:
+  typedef std::input_iterator_tag iterator_category;
+  typedef typename F::result_type value_type;
+  typedef const value_type&       reference;
+  typedef const value_type*       pointer;
+  typedef void                    difference_type;
+
+  explicit 
+  generator_iterator(RandomGenerator& gen, const F& f = F()) 
+    : f(f), gen(&gen) 
+  { 
+    value = this->f(gen); 
+  }
+
+  reference operator*() const  { return value; }
+  pointer   operator->() const { return &value; }
+
+  generator_iterator& operator++()
+  {
+    value = f(*gen);
+    return *this;
+  }
+
+  generator_iterator operator++(int)
+  {
+    generator_iterator temp(*this);
+    ++(*this);
+    return temp;
+  }
+
+  bool operator==(const generator_iterator& other) const
+  { return f == other.f; }
+
+  bool operator!=(const generator_iterator& other) const
+  { return !(*this == other); }
+
+private:
+  F f;
+  RandomGenerator* gen;
+  value_type value;
+};
+
+template<typename F, typename RandomGenerator>
+inline generator_iterator<F, RandomGenerator> 
+make_generator_iterator( RandomGenerator& gen, const F& f)
+{ return generator_iterator<F, RandomGenerator>(gen, f); }
+
+/****************************************************************************
+ * Edge Property                                                            *
+ ****************************************************************************/
+typedef int weight_type;
+
+struct WeightedEdge {
+  WeightedEdge(weight_type weight = 0) : weight(weight) { }
+  
+  weight_type weight;
+
+  template<typename Archiver>
+  void serialize(Archiver& ar, const unsigned int /*version*/)
+  {
+    ar & weight;
+  }
+};
+
+/****************************************************************************
+ * Algorithm Tests                                                          *
+ ****************************************************************************/
+template <typename Graph>
+void test_directed_sequential_algorithms(const Graph& g)
+{ }
+
+template <typename Graph>
+void test_undirected_sequential_algorithms(const Graph& g)
+{ 
+  std::vector<unsigned int> componentS(num_vertices(g));
+  typedef iterator_property_map<
+      std::vector<unsigned int>::iterator, 
+      typename property_map<Graph, vertex_index_t>::type> 
+    ComponentMap;
+  ComponentMap component(componentS.begin(), get(vertex_index, g));
+
+  time_type start = get_time();
+  unsigned int num_components = connected_components(g, component);
+  time_type end = get_time();
+  
+  std::cerr << "    Sequential connected Components time = " 
+            << print_time(end - start) << " seconds.\n" 
+            << "    " << num_components << " components identified\n";
+}
+
+template <typename Graph, typename EdgeWeightMap>
+void test_directed_csr_only_algorithms(const Graph& g, EdgeWeightMap weight,
+                                       typename graph_traits<Graph>::vertices_size_type num_sources,
+                                       typename property_traits<EdgeWeightMap>::value_type C)
+{
+  typedef typename graph_traits<Graph>::vertex_descriptor vertex_descriptor;
+  typedef typename graph_traits<Graph>::vertex_iterator vertex_iterator;
+  typedef typename graph_traits<Graph>::vertices_size_type vertices_size_type;
+  typedef typename graph_traits<Graph>::edges_size_type edges_size_type;
+
+  typedef typename boost::graph::parallel::process_group_type<Graph>::type 
+    process_group_type;
+
+  process_group_type pg = process_group(g);
+  typename process_group_type::process_id_type id = process_id(pg);
+  typename process_group_type::process_size_type p = num_processes(pg);    
+
+  vertices_size_type n = num_vertices(g);
+  n = boost::parallel::all_reduce(pg, n, std::plus<vertices_size_type>());
+
+  edges_size_type m = num_edges(g);
+  m = boost::parallel::all_reduce(pg, m, std::plus<edges_size_type>());
+
+  //
+  // Betweenness Centrality (Approximate)
+  //
+  queue<vertex_descriptor> delta_stepping_vertices;
+
+  {
+    // Distributed Centrality Map
+    typedef typename property_map<Graph, vertex_index_t>::const_type IndexMap;
+    typedef iterator_property_map<std::vector<int>::iterator, IndexMap> CentralityMap;
+    
+    std::vector<int> centralityS(num_vertices(g), 0);
+    CentralityMap centrality(centralityS.begin(), get(vertex_index, g));
+
+    // Calculate number of vertices of degree 0
+    vertices_size_type n0 = 0;
+    BGL_FORALL_VERTICES_T(v, g, Graph) {
+      if (out_degree(v, g) == 0) n0++;
+    }
+    n0 = boost::parallel::all_reduce(pg, n0, std::plus<vertices_size_type>());
+
+    queue<vertex_descriptor> sources;
+
+    {
+      assert(num_sources >= p);  // Don't feel like adding a special case for num_sources < p
+
+      minstd_rand gen;
+      uniform_int<vertices_size_type> rand_vertex(0, num_vertices(g) - 1);
+      std::vector<vertex_descriptor> all_sources, local_sources;
+      vertices_size_type local_vertices = vertices_size_type(floor((double)num_sources / p));
+      local_vertices += (id < (num_sources - (p * local_vertices)) ? 1 : 0); 
+
+      while (local_vertices > 0) {
+        vertex_iterator iter = vertices(g).first;
+        std::advance(iter, rand_vertex(gen));
+        
+        if (out_degree(*iter, g) != 0
+            && std::find(local_sources.begin(), local_sources.end(), *iter) == local_sources.end()) {
+          local_sources.push_back(*iter);
+          --local_vertices;
+        }
+      }
+      all_gather(pg, local_sources.begin(), local_sources.end(), all_sources);
+      std::sort(all_sources.begin(), all_sources.end());
+      for (typename std::vector<vertex_descriptor>::iterator iter = all_sources.begin(); 
+           iter != all_sources.end(); ++iter) {
+        sources.push(*iter);
+        delta_stepping_vertices.push(*iter);
+      }
+    }
+
+    // NOTE: The delta below assumes uniform edge weight distribution
+    time_type start = get_time();
+    brandes_betweenness_centrality(g, buffer(sources).weight_map(weight).
+                                   centrality_map(centrality).lookahead((m / n) * (C / 2)));
+    time_type end = get_time();
+    
+    edges_size_type exactTEPs = edges_size_type(floor(7 * n* (n - n0) / (end - start)));
+    
+    if (id == 0)
+      std::cerr << "    Betweenness Centrality Approximate (" << num_sources << " sources) = " 
+                << print_time(end - start) << " (" << exactTEPs << " TEPs)\n";
+  }
+
+  //
+  // Delta stepping performance (to compare to SSSP inside BC
+  //
+  if (false) {
+    typedef typename property_map<Graph, vertex_index_t>::const_type IndexMap;
+    typedef iterator_property_map<std::vector<int>::iterator, IndexMap> DistanceMap;
+    
+    std::vector<int> distanceS(num_vertices(g), 0);
+    DistanceMap distance(distanceS.begin(), get(vertex_index, g));
+
+    while(!delta_stepping_vertices.empty()) {
+
+      time_type start = get_time();
+      delta_stepping_shortest_paths(g, delta_stepping_vertices.top(), dummy_property_map(),
+                                    distance, weight);
+      time_type end = get_time();
+  
+      delta_stepping_vertices.pop();
+      distance.reset();
+    
+      if (id == 0)
+        std::cerr << "    Delta-stepping shortest paths = " << print_time(end - start) 
+                  << std::endl;
+    }
+  }
+
+}
+
+template <typename Graph>
+void test_directed_algorithms(const Graph& g)
+{
+}
+
+template <typename Graph>
+void test_undirected_algorithms(const Graph& g)
+{
+  typedef typename boost::graph::parallel::process_group_type<Graph>::type 
+    process_group_type;
+
+  process_group_type pg = process_group(g);
+  typename process_group_type::process_id_type id = process_id(pg);
+  typename process_group_type::process_size_type p = num_processes(pg);    
+
+
+  // Connected Components
+  std::vector<unsigned int> local_components_vec(num_vertices(g));
+  typedef iterator_property_map<
+      std::vector<unsigned int>::iterator, 
+      typename property_map<Graph, vertex_index_t>::type> 
+    ComponentMap;
+  ComponentMap component(local_components_vec.begin(), get(vertex_index, g));
+
+  int num_components;
+
+  time_type start = get_time();
+  num_components = connected_components(g, component);
+  time_type end = get_time();
+  if (id == 0) 
+    std::cerr << "    Connected Components time = " << print_time(end - start) 
+              << " seconds.\n" 
+              << "    " << num_components << " components identified\n";
+
+  start = get_time();
+  num_components = boost::graph::distributed::connected_components_ps(g, component);
+  end = get_time();
+  if (id == 0) 
+    std::cerr << "    Connected Components (parallel search) time = " 
+              << print_time(end - start) << " seconds.\n"
+              << "    " << num_components << " components identified\n";
+}
+
+/****************************************************************************
+ * Graph Type Tests                                                         *
+ ****************************************************************************/
+
+// TODO: Re-seed PRNG before sequential tests to get the same graph as the 
+// distributed case?
+
+//
+// Compressed Sparse Row
+//
+
+// R-MAT 
+template <typename ProcessGroup, typename RandomGenerator, typename Distribution>
+void test_csr(const ProcessGroup& pg, RandomGenerator& gen, Distribution& distrib,
+              bool sequential_tests, size_t N, size_t M, size_t C, 
+              double a, double b, double c, double d, size_t num_sources,
+              rmat_edge_distribution_tag)
+{
+  if (process_id(pg) == 0)
+    std::cerr << "  R-MAT\n";
+
+  typedef compressed_sparse_row_graph<directedS, no_property, WeightedEdge, no_property,
+                                      distributedS<mpi_process_group> > Graph;
+
+  Graph g(sorted_rmat_iterator<RandomGenerator, Graph>(gen, N, M, a, b, c, d),
+          sorted_rmat_iterator<RandomGenerator, Graph>(),
+          make_generator_iterator(gen, uniform_int<int>(1, C)),
+          N, pg, distrib);
+  
+  test_directed_algorithms(g);
+  test_directed_csr_only_algorithms(g, get(&WeightedEdge::weight, g), num_sources, C);
+
+  if (sequential_tests && process_id(pg) == 0) {
+    typedef compressed_sparse_row_graph<directedS, no_property, WeightedEdge>
+      seqGraph;
+    
+    seqGraph sg(edges_are_sorted,
+                sorted_rmat_iterator<RandomGenerator, seqGraph>(gen, N, M, a, b, c, d),
+                sorted_rmat_iterator<RandomGenerator, seqGraph>(),
+                make_generator_iterator(gen, uniform_int<int>(1, C)),
+                N);
+    
+    test_directed_sequential_algorithms(sg);
+  }
+}
+
+// R-MAT with unique edges
+template <typename ProcessGroup, typename RandomGenerator, typename Distribution>
+void test_csr(const ProcessGroup& pg, RandomGenerator& gen, Distribution& distrib,
+              bool sequential_tests, size_t N, size_t M, size_t C,  
+              double a, double b, double c, double d, size_t num_sources,
+              rmat_unique_edge_distribution_tag)
+{
+  if (process_id(pg) == 0)
+    std::cerr << "  R-MAT - unique\n";
+
+  typedef compressed_sparse_row_graph<directedS, no_property, WeightedEdge, no_property,
+                                      distributedS<mpi_process_group> > Graph;
+  
+  // Last boolean parameter makes R-MAT bidirectional
+  Graph g(sorted_unique_rmat_iterator<RandomGenerator, Graph>(gen, N, M, a, b, c, d),
+          sorted_unique_rmat_iterator<RandomGenerator, Graph>(),
+          make_generator_iterator(gen, uniform_int<int>(1, C)),
+          N, pg, distrib);
+  
+  test_directed_algorithms(g);
+  test_directed_csr_only_algorithms(g, get(&WeightedEdge::weight, g), num_sources, C);
+
+  if (sequential_tests && process_id(pg) == 0) { 
+    typedef compressed_sparse_row_graph<directedS, no_property, WeightedEdge>
+      seqGraph;
+    
+    seqGraph sg(edges_are_sorted,
+                sorted_unique_rmat_iterator<RandomGenerator, seqGraph>(gen, N, M, a, b, c, d),
+                sorted_unique_rmat_iterator<RandomGenerator, seqGraph>(),
+                make_generator_iterator(gen, uniform_int<int>(1, C)),
+                N);
+    
+    test_directed_sequential_algorithms(sg);
+  }
+}
+
+// Erdos-Renyi
+template <typename ProcessGroup, typename RandomGenerator, typename Distribution>
+void test_csr(const ProcessGroup& pg, RandomGenerator& gen, Distribution& distrib,
+              bool sequential_tests, size_t N, size_t M, size_t C, size_t num_sources)
+{
+  if (process_id(pg) == 0)
+    std::cerr << "  Erdos-Renyi\n";
+
+  double _p = static_cast<double>(M) / (N*N);
+
+  typedef compressed_sparse_row_graph<directedS, no_property, WeightedEdge, no_property,
+                                      distributedS<mpi_process_group> > Graph;
+    
+  Graph g(sorted_erdos_renyi_iterator<RandomGenerator, Graph>(gen, N, _p/2),
+          sorted_erdos_renyi_iterator<RandomGenerator, Graph>(),
+          make_generator_iterator(gen, uniform_int<int>(1, C)),
+          N, pg, distrib);
+  
+  test_directed_algorithms(g);
+  test_directed_csr_only_algorithms(g, get(&WeightedEdge::weight, g), num_sources, C);
+
+  if (sequential_tests && process_id(pg) == 0) {
+    typedef compressed_sparse_row_graph<directedS, no_property, WeightedEdge>
+      seqGraph;
+    
+    seqGraph sg(edges_are_sorted,
+                sorted_erdos_renyi_iterator<RandomGenerator, seqGraph>(gen, N, _p/2),
+                sorted_erdos_renyi_iterator<RandomGenerator, seqGraph>(),
+                make_generator_iterator(gen, uniform_int<int>(1, C)),
+                N);
+    
+    test_directed_sequential_algorithms(sg);
+  }
+}
+
+// Small World
+template <typename ProcessGroup, typename RandomGenerator, typename Distribution>
+void test_csr(const ProcessGroup& pg, RandomGenerator& gen, Distribution& distrib,
+              bool sequential_tests, size_t N, size_t M, size_t C, double p, 
+              size_t num_sources)
+{
+  if (process_id(pg) == 0)
+    std::cerr << "  Small-World\n";
+
+  int k = M / N;
+
+  typedef compressed_sparse_row_graph<directedS, no_property, WeightedEdge, no_property,
+                                      distributedS<mpi_process_group> > Graph;
+
+  Graph g(small_world_iterator<RandomGenerator, Graph>(gen, N, k, p),
+          small_world_iterator<RandomGenerator, Graph>(),
+          make_generator_iterator(gen, uniform_int<int>(1, C)),
+          N, pg, distrib);
+  
+  test_directed_algorithms(g);
+  test_directed_csr_only_algorithms(g, get(&WeightedEdge::weight, g), num_sources, C);
+
+  if (sequential_tests && process_id(pg) == 0) {
+    typedef compressed_sparse_row_graph<directedS, no_property, WeightedEdge>
+      seqGraph;
+    
+    seqGraph sg(edges_are_sorted,
+                small_world_iterator<RandomGenerator, seqGraph>(gen, N, k, p),
+                small_world_iterator<RandomGenerator, seqGraph>(),
+                make_generator_iterator(gen, uniform_int<int>(1, C)),
+                N);
+    
+    test_directed_sequential_algorithms(sg);
+  }
+}
+
+//
+// Adjacency List
+//
+
+// R-MAT 
+template <typename ProcessGroup, typename RandomGenerator, typename Distribution>
+void test_adjacency_list(const ProcessGroup& pg, RandomGenerator& gen, Distribution& distrib,
+                         bool sequential_tests, size_t N, size_t M, size_t C, 
+                         double a, double b, double c, double d,
+                         rmat_edge_distribution_tag)
+{
+  if (process_id(pg) == 0)
+    std::cerr << "R-MAT\n";
+
+  {
+    typedef adjacency_list<vecS, distributedS<mpi_process_group, vecS>,
+                           directedS, no_property, WeightedEdge> Graph;
+    
+    Graph g(rmat_iterator<RandomGenerator, Graph>(gen, N, M, a, b, c, d),
+            rmat_iterator<RandomGenerator, Graph>(),
+            make_generator_iterator(gen, uniform_int<int>(1, C)),
+            N, pg, distrib);
+
+    test_directed_algorithms(g);
+  }
+
+  {
+    typedef adjacency_list<vecS, distributedS<mpi_process_group, vecS>,
+                           undirectedS, no_property, WeightedEdge> Graph;
+    
+    Graph g(rmat_iterator<RandomGenerator, Graph>(gen, N, M, a, b, c, d),
+            rmat_iterator<RandomGenerator, Graph>(),
+            make_generator_iterator(gen, uniform_int<int>(1, C)),
+            N, pg, distrib);
+
+    test_undirected_algorithms(g);
+  }
+
+  if (sequential_tests && process_id(pg) == 0) {
+    {
+      typedef adjacency_list<vecS, vecS, directedS, no_property, property<edge_weight_t, int> > 
+        seqGraph;
+      
+      seqGraph sg(rmat_iterator<RandomGenerator, seqGraph>(gen, N, M, a, b, c, d),
+                  rmat_iterator<RandomGenerator, seqGraph>(),
+                  make_generator_iterator(gen, uniform_int<int>(1, C)),
+                  N);
+
+      test_directed_sequential_algorithms(sg);
+    }
+
+    {
+      typedef adjacency_list<vecS, vecS, undirectedS, no_property, property<edge_weight_t, int> > 
+        seqGraph;
+      
+      seqGraph sg(rmat_iterator<RandomGenerator, seqGraph>(gen, N, M, a, b, c, d),
+                  rmat_iterator<RandomGenerator, seqGraph>(),
+                  make_generator_iterator(gen, uniform_int<int>(1, C)),
+                  N);
+
+      test_undirected_sequential_algorithms(sg);
+    }
+  }
+}
+
+// R-MAT with unique edges
+template <typename ProcessGroup, typename RandomGenerator, typename Distribution>
+void test_adjacency_list(const ProcessGroup& pg, RandomGenerator& gen, Distribution& distrib,
+                         bool sequential_tests, size_t N, size_t M, size_t C,  
+                         double a, double b, double c, double d,
+                         rmat_unique_edge_distribution_tag)
+{
+  if (process_id(pg) == 0)
+    std::cerr << "  R-MAT - unique\n";
+
+  {
+    typedef adjacency_list<vecS, distributedS<mpi_process_group, vecS>,
+                           directedS, no_property, WeightedEdge> Graph;
+
+    Graph g(sorted_unique_rmat_iterator<RandomGenerator, Graph>(gen, N, M, a, b, c, d),
+            sorted_unique_rmat_iterator<RandomGenerator, Graph>(),
+            make_generator_iterator(gen, uniform_int<int>(1, C)),
+            N, pg, distrib);
+    
+    test_directed_algorithms(g);
+  }
+
+  {
+    typedef adjacency_list<vecS, distributedS<mpi_process_group, vecS>,
+                           undirectedS, no_property, WeightedEdge> Graph;
+
+    Graph g(sorted_unique_rmat_iterator<RandomGenerator, Graph>(gen, N, M, a, b, c, d),
+            sorted_unique_rmat_iterator<RandomGenerator, Graph>(),
+            make_generator_iterator(gen, uniform_int<int>(1, C)),
+            N, pg, distrib);
+    
+    test_undirected_algorithms(g);
+  }
+
+  if (sequential_tests && process_id(pg) == 0) {
+    {
+      typedef adjacency_list<vecS, vecS, directedS, no_property, property<edge_weight_t, int> > 
+        seqGraph;
+      
+      seqGraph sg(sorted_unique_rmat_iterator<RandomGenerator, seqGraph>(gen, N, M, a, b, c, d),
+                  sorted_unique_rmat_iterator<RandomGenerator, seqGraph>(),
+                  make_generator_iterator(gen, uniform_int<int>(1, C)),
+                  N);
+
+      test_directed_sequential_algorithms(sg);
+    }
+
+    {
+      typedef adjacency_list<vecS, vecS, undirectedS, no_property, property<edge_weight_t, int> > 
+        seqGraph;
+      
+      seqGraph sg(sorted_unique_rmat_iterator<RandomGenerator, seqGraph>(gen, N, M, a, b, c, d),
+                  sorted_unique_rmat_iterator<RandomGenerator, seqGraph>(),
+                  make_generator_iterator(gen, uniform_int<int>(1, C)),
+                  N);
+
+      test_undirected_sequential_algorithms(sg);
+    }
+  }
+}
+
+// Erdos-Renyi
+template <typename ProcessGroup, typename RandomGenerator, typename Distribution>
+void test_adjacency_list(const ProcessGroup& pg, RandomGenerator& gen, Distribution& distrib,
+                         bool sequential_tests, size_t N, size_t M, size_t C)
+{
+  if (process_id(pg) == 0)
+    std::cerr << "  Erdos-Renyi\n";
+
+  double _p = static_cast<double>(M) / N*N;
+
+  {
+    typedef adjacency_list<vecS, distributedS<mpi_process_group, vecS>,
+                           directedS, no_property, WeightedEdge> Graph;
+
+    Graph g(erdos_renyi_iterator<RandomGenerator, Graph>(gen, N, _p/2),
+            erdos_renyi_iterator<RandomGenerator, Graph>(),
+            make_generator_iterator(gen, uniform_int<int>(1, C)),
+            N, pg, distrib);
+    
+    test_directed_algorithms(g);
+  }
+
+  {
+    typedef adjacency_list<vecS, distributedS<mpi_process_group, vecS>,
+                           undirectedS, no_property, WeightedEdge> Graph;
+
+    Graph g(erdos_renyi_iterator<RandomGenerator, Graph>(gen, N, _p/2),
+            erdos_renyi_iterator<RandomGenerator, Graph>(),
+            make_generator_iterator(gen, uniform_int<int>(1, C)),
+            N, pg, distrib);
+    
+    test_undirected_algorithms(g);
+  }
+
+  if (sequential_tests && process_id(pg) == 0) {
+    {
+      typedef adjacency_list<vecS, vecS, directedS, no_property, property<edge_weight_t, int> > 
+        seqGraph;
+
+      seqGraph sg(erdos_renyi_iterator<RandomGenerator, seqGraph>(gen, N, _p/2),
+                  erdos_renyi_iterator<RandomGenerator, seqGraph>(),
+                  make_generator_iterator(gen, uniform_int<int>(1, C)),
+                  N);
+      
+      test_directed_sequential_algorithms(sg);
+    }
+
+    {
+      typedef adjacency_list<vecS, vecS, undirectedS, no_property, property<edge_weight_t, int> > 
+        seqGraph;
+
+      seqGraph sg(erdos_renyi_iterator<RandomGenerator, seqGraph>(gen, N, _p/2),
+                  erdos_renyi_iterator<RandomGenerator, seqGraph>(),
+                  make_generator_iterator(gen, uniform_int<int>(1, C)),
+                  N);
+      
+      test_undirected_sequential_algorithms(sg);
+    }
+  }
+}
+
+// Small World
+template <typename ProcessGroup, typename RandomGenerator, typename Distribution>
+void test_adjacency_list(const ProcessGroup& pg, RandomGenerator& gen, Distribution& distrib,
+                         bool sequential_tests, size_t N, size_t M, size_t C, double p)
+{
+  if (process_id(pg) == 0)
+    std::cerr << "  Small-World\n";
+
+  int k = M / N;
+
+  {
+    typedef adjacency_list<vecS, distributedS<mpi_process_group, vecS>,
+                           directedS, no_property, WeightedEdge> Graph;
+
+    Graph g(small_world_iterator<RandomGenerator, Graph>(gen, N, k, p),
+            small_world_iterator<RandomGenerator, Graph>(),
+            make_generator_iterator(gen, uniform_int<int>(1, C)),
+            N, pg, distrib);
+
+    test_directed_algorithms(g);
+  }
+
+  {
+    typedef adjacency_list<vecS, distributedS<mpi_process_group, vecS>,
+                           undirectedS, no_property, WeightedEdge> Graph;
+
+    Graph g(small_world_iterator<RandomGenerator, Graph>(gen, N, k, p),
+            small_world_iterator<RandomGenerator, Graph>(),
+            make_generator_iterator(gen, uniform_int<int>(1, C)),
+            N, pg, distrib);
+
+    test_undirected_algorithms(g);
+  }
+
+  if (sequential_tests && process_id(pg) == 0) {
+    {
+      typedef adjacency_list<vecS, vecS, directedS, no_property, property<edge_weight_t, int> > 
+        seqGraph;
+      
+      seqGraph sg(small_world_iterator<RandomGenerator, seqGraph>(gen, N, k, p),
+                  small_world_iterator<RandomGenerator, seqGraph>(),
+                  make_generator_iterator(gen, uniform_int<int>(1, C)),
+                  N);
+      
+      test_directed_sequential_algorithms(sg);
+    }
+
+    {
+      typedef adjacency_list<vecS, vecS, undirectedS, no_property, property<edge_weight_t, int> > 
+        seqGraph;
+      
+      seqGraph sg(small_world_iterator<RandomGenerator, seqGraph>(gen, N, k, p),
+                  small_world_iterator<RandomGenerator, seqGraph>(),
+                  make_generator_iterator(gen, uniform_int<int>(1, C)),
+                  N);
+      
+      test_undirected_sequential_algorithms(sg);
+    }
+  }
+}
+
+void usage()
+{
+  std::cerr << "Algorithm Performance Test\n"
+            << "Usage : algorithm_performance [options]\n\n"
+            << "Options are:\n"
+            << "\t--vertices v\t\t\tNumber of vertices in the graph\n"
+            << "\t--edges v\t\t\tNumber of edges in the graph\n"
+            << "\t--seed s\t\t\tSeed for synchronized random number generator\n"
+            << "\t--max-weight miw\t\tMaximum integer edge weight\n"
+            << "\t--rewire-probability\t\tRewire-probabitility (p) for small-world graphs\n"
+            << "\t--dot\t\t\t\tEmit a dot file containing the graph\n"
+            << "\t--verify\t\t\tVerify result\n"
+            << "\t--degree-dist\t\t\tOutput degree distribution of graph\n"
+            << "\t--sequential-graph\t\tRun sequential graph tests\n"
+            << "\t--erdos-renyi\t\t\tRun tests on Erdos-Renyi graphs\n"   
+            << "\t--small-world\t\t\tRun tests on Small World graphs\n"   
+            << "\t--rmat\t\t\t\tRun tests on R-MAT graphs\n"   
+            << "\t--rmat-unique\t\t\tRun tests on R-MAT graphs with no duplicate edges\n"
+            << "\t--csr <bool>\t\t\tRun tests using CSR graphs\n"   
+            << "\t--adjacency-list <bool>\t\tRun tests using Adjacency List graphs\n";   
+}
+
+
+int test_main(int argc, char* argv[])
+{
+  mpi::environment env(argc, argv);
+
+  rand48 gen;
+  
+  // Default args
+  size_t n = 100, 
+    m = 8*n, 
+    c = 100,
+    num_sources = 32,
+    num_headers = 16 * 1024,
+    batch_buffer_size = 1024 * 1024;
+  uint64_t seed = 1;
+  bool emit_dot_file = false, 
+    verify = false,
+    sequential_graph = false,
+    show_degree_dist = false,
+    erdos_renyi = false,
+    small_world = false,
+    rmat = false,
+    rmat_unique = false,
+    csr = false,
+    adj_list = false;
+  double p = 0.1,
+    rmat_a = 0.5, 
+    rmat_b = 0.25,
+    rmat_c = 0.1,
+    rmat_d = 0.15; 
+
+  // Parse args
+  for (int i = 1; i < argc; ++i) {
+    std::string arg = argv[i];
+
+    if (arg == "--vertices")
+      n = boost::lexical_cast<size_t>( argv[i+1] );
+
+    if (arg == "--seed")
+      seed = boost::lexical_cast<uint64_t>( argv[i+1] );
+
+    if (arg == "--edges")
+      m = boost::lexical_cast<size_t>( argv[i+1] );
+
+    if (arg == "--max-weight")
+      c = boost::lexical_cast<int>( argv[i+1] );
+
+    if (arg == "--batch-buffer-size") {
+      batch_buffer_size = boost::lexical_cast<size_t>( argv[i+1] );
+      num_headers = batch_buffer_size / 16;
+      num_headers = num_headers > 0 ? num_headers : 1;
+    }
+
+    if (arg == "--rewire-probability")
+      p = boost::lexical_cast<double>( argv[i+1] );
+
+    if (arg == "--num-sources")
+      num_sources = boost::lexical_cast<size_t>( argv[i + 1] );
+
+    if (arg == "--erdos-renyi")
+      erdos_renyi = true;
+
+    if (arg == "--small-world")
+      small_world = true;
+
+    if (arg == "--rmat")
+      rmat = true;
+
+    if (arg == "--rmat-unique")
+      rmat_unique = true;
+
+    if (arg == "--dot")
+      emit_dot_file = true;
+
+    if (arg == "--verify")
+      verify = true;
+
+    if (arg == "--degree-dist")
+      show_degree_dist = true;
+
+    if (arg == "--sequential-graph")
+      sequential_graph = true;
+
+    if (arg == "--csr")
+      csr = std::string(argv[i+1]) == "true";
+
+    if (arg == "--adjacency-list")
+      adj_list = std::string(argv[i+1]) == "true";
+  }
+
+  mpi_process_group pg(num_headers, batch_buffer_size);
+
+  if (argc == 1) {
+    if (process_id(pg) == 0)
+      usage();
+    exit(-1);
+  }
+
+  gen.seed(seed);
+
+  parallel::variant_distribution<mpi_process_group> distrib 
+    = parallel::block(pg, n);
+
+  if (csr) {
+    if (process_id(pg) == 0)
+      std::cerr << "CSR Graph Tests\n";
+
+    if (erdos_renyi)
+      test_csr(pg, gen, distrib, sequential_graph, n, m, c, num_sources);
+    if (small_world)
+      test_csr(pg, gen, distrib, sequential_graph, n, m, c, p, num_sources);
+    if (rmat)
+      test_csr(pg, gen, distrib, sequential_graph, n, m, c, 
+               rmat_a, rmat_b, rmat_c, rmat_d, num_sources, rmat_edge_distribution_tag());
+    if (rmat_unique)
+      test_csr(pg, gen, distrib, sequential_graph, n, m, c, 
+               rmat_a, rmat_b, rmat_c, rmat_d, num_sources, rmat_unique_edge_distribution_tag());
+  }
+
+  gen.seed(seed); // DEBUG 
+
+  if (adj_list) {
+    if (process_id(pg) == 0)
+      std::cerr << "Adjacency List Graph Tests\n";
+
+    if (erdos_renyi)
+      test_adjacency_list(pg, gen, distrib, sequential_graph, n, m, c);
+    if (small_world)
+      test_adjacency_list(pg, gen, distrib, sequential_graph, n, m, c, p);
+    if (rmat)
+      test_adjacency_list(pg, gen, distrib, sequential_graph, n, m, c, 
+                          rmat_a, rmat_b, rmat_c, rmat_d, rmat_edge_distribution_tag());
+    if (rmat_unique)
+      test_adjacency_list(pg, gen, distrib, sequential_graph, n, m, c, 
+                          rmat_a, rmat_b, rmat_c, rmat_d, rmat_unique_edge_distribution_tag());
+  }
+
+  return 0;
+}
diff --git a/src/boost/libs/graph_parallel/test/distributed_adjacency_list_test.cpp b/src/boost/libs/graph_parallel/test/distributed_adjacency_list_test.cpp
new file mode 100644
index 00000000..cc7ce8cf
--- /dev/null
+++ b/src/boost/libs/graph_parallel/test/distributed_adjacency_list_test.cpp
@@ -0,0 +1,284 @@
+// Copyright (C) 2004-2008 The Trustees of Indiana University.
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+//  Authors: Douglas Gregor
+//           Andrew Lumsdaine
+
+#include <boost/graph/use_mpi.hpp>
+#include <boost/config.hpp>
+#include <boost/throw_exception.hpp>
+#include <boost/graph/distributed/adjacency_list.hpp>
+#include <boost/graph/distributed/mpi_process_group.hpp>
+#include <boost/graph/distributed/local_subgraph.hpp>
+#include <boost/graph/parallel/distribution.hpp>
+#include <iostream>
+#include <cassert>
+#include <boost/test/minimal.hpp>
+
+#ifdef BOOST_NO_EXCEPTIONS
+void
+boost::throw_exception(std::exception const& ex)
+{
+    std::cout << ex.what() << std::endl;
+    abort();
+}
+#endif
+
+using namespace boost;
+using boost::graph::distributed::mpi_process_group;
+
+template<typename Graph>
+struct never
+{
+  typedef typename graph_traits<Graph>::edge_descriptor argument_type;
+  typedef bool result_type;
+
+  result_type operator()(argument_type) { return false; }
+};
+
+int test_main(int argc, char** argv)
+{
+  boost::mpi::environment env(argc, argv);
+
+  mpi_process_group pg;
+  parallel::block dist(pg, 20);
+
+  typedef adjacency_list<listS, distributedS<mpi_process_group, vecS>,
+                         bidirectionalS> Graph1;
+  typedef adjacency_list<listS, distributedS<mpi_process_group, vecS>,
+                         directedS> Graph2;
+  typedef adjacency_list<listS, distributedS<mpi_process_group, vecS>,
+                         undirectedS> Graph3;
+
+  if (num_processes(pg) > 20) return -1;
+
+  if (process_id(pg) == 0) std::cout << "Graph 1------------------\n";
+
+  std::cout << "Processor #" << process_id(pg) << ": "
+            << dist.block_size(20) << " vertices." << std::endl;
+  {
+    Graph1 g1(20);
+
+    //    if (process_id(pg) == num_processes(pg)() - 1)
+    //      add_vertex(g1);
+
+    graph_traits<Graph1>::vertex_iterator v, v_end;
+    int counter = 0;
+    for (boost::tie(v, v_end) = vertices(g1); v != v_end; ++v) {
+      std::cout << "Processor #" << process_id(pg) << ": vertex " << ++counter
+                << std::endl;
+
+      out_edges(*v, g1);
+      out_degree(*v, g1);
+      in_edges(*v, g1);
+      in_degree(*v, g1);
+
+      graph_traits<Graph1>::vertex_descriptor other = *v;
+      other.owner = (other.owner + 1) % num_processes(pg);
+      other.local = 0;
+      add_edge(*v, other, g1);
+
+      std::cout << "Adding edge from processor " << process_id(pg)
+                << " to processor " << other.owner << std::endl;
+    }
+
+    synchronize(g1);
+    assert((std::size_t)std::distance(edges(g1).first, edges(g1).second) == num_vertices(g1));
+
+    if (num_vertices(g1) >= 2) {
+      graph_traits<Graph1>::vertex_iterator vi = vertices(g1).first;
+      graph_traits<Graph1>::vertex_descriptor u = *vi++;
+      graph_traits<Graph1>::vertex_descriptor v = *vi++;
+      add_edge(u, v, g1);
+      assert(out_degree(u, g1) == 2);
+      assert(in_degree(v, g1) == 1);
+    }
+
+    int prior_processor = (process_id(pg) + num_processes(pg) - 1)
+      % num_processes(pg);
+    const int n = 20;
+    std::size_t vertices_in_prior = (n / num_processes(pg))
+      + (n % num_processes(pg) > prior_processor? 1 : 0);
+
+    graph_traits<Graph1>::vertex_descriptor u = *vertices(g1).first;
+    if (in_degree(u, g1) != vertices_in_prior) {
+      std::cout << "Processor #" << process_id(pg) << ": " << in_degree(u, g1)
+                << " vs. " << vertices_in_prior << std::endl;
+    }
+    assert(in_degree(u, g1) == vertices_in_prior);
+    std::cout << "Processor #" << process_id(pg) << ": " << num_edges(g1)
+              << " edges.\n";
+
+    local_subgraph<Graph1> local_g1(g1);
+    edges(local_g1);
+    vertices(local_g1);
+    if (num_vertices(local_g1) > 0) {
+      out_edges(*vertices(local_g1).first, local_g1);
+      in_edges(*vertices(local_g1).first, local_g1);
+      adjacent_vertices(*vertices(local_g1).first, local_g1);
+      if (false) {
+        remove_out_edge_if(*vertices(g1).first, never<Graph1>(), g1);
+        remove_in_edge_if(*vertices(g1).first, never<Graph1>(), g1);
+        clear_out_edges(*vertices(g1).first, g1);
+        clear_in_edges(*vertices(g1).first, g1);
+        clear_vertex(*vertices(g1).first, g1);
+        remove_vertex(*vertices(g1).first, g1);
+      }
+    }
+    remove_edge_if(never<Graph1>(), g1);
+  }
+
+
+  if (process_id(pg) == 0) std::cout << "Graph 2------------------\n";
+
+  {
+    Graph2 g2(20);
+
+    if (process_id(pg) == num_processes(pg) - 1)
+      add_vertex(g2);
+
+    graph_traits<Graph2>::vertex_iterator v, v_end;
+    int counter = 0;
+    for (boost::tie(v, v_end) = vertices(g2); v != v_end; ++v) {
+      std::cout << "Processor #" << process_id(pg) << ": vertex " << ++counter
+                << std::endl;
+
+      out_edges(*v, g2);
+    }
+
+    synchronize(g2);
+
+    if (num_vertices(g2) >= 2) {
+      graph_traits<Graph2>::vertex_iterator vi = vertices(g2).first;
+      graph_traits<Graph2>::vertex_descriptor u = *vi++;
+      graph_traits<Graph2>::vertex_descriptor v = *vi++;
+      add_edge(u, v, g2);
+      assert(out_degree(u, g2) == 1);
+      assert(*adjacent_vertices(u, g2).first == v);
+      std::cout << "Processor #" << process_id(pg) << ": " << num_edges(g2)
+                << " edges.\n";
+      assert(std::distance(edges(g2).first, edges(g2).second) == 1);
+
+    }
+    synchronize(g2);
+
+    local_subgraph<Graph2> local_g2(g2);
+    edges(local_g2);
+    vertices(local_g2);
+    if (num_vertices(local_g2) > 0) {
+      out_edges(*vertices(local_g2).first, local_g2);
+      adjacent_vertices(*vertices(local_g2).first, local_g2);
+      remove_out_edge_if(*vertices(g2).first, never<Graph2>(), g2);
+      clear_out_edges(*vertices(g2).first, g2);
+      remove_vertex(*vertices(g2).first, g2);
+    }
+    remove_edge_if(never<Graph2>(), g2);
+  }
+
+  if (process_id(pg) == 0) std::cout << "Graph 3------------------\n";
+
+  {
+    Graph3 g3(20);
+
+    //    if (process_id(pg) == num_processes(pg) - 1)
+    //      add_vertex(g3);
+
+    graph_traits<Graph3>::vertex_iterator v, v_end;
+    int counter = 0;
+    for (boost::tie(v, v_end) = vertices(g3); v != v_end; ++v) {
+      std::cout << "Processor #" << process_id(pg) << ": vertex " << ++counter
+                << std::endl;
+
+      out_edges(*v, g3);
+      out_degree(*v, g3);
+      in_edges(*v, g3);
+      in_degree(*v, g3);
+    }
+
+    graph_traits<Graph3>::vertices_size_type added_edges = 0;
+    if (num_vertices(g3) >= 2) {
+      graph_traits<Graph3>::vertex_iterator vi = vertices(g3).first;
+      graph_traits<Graph3>::vertex_descriptor u = *vi++;
+      graph_traits<Graph3>::vertex_descriptor v = *vi++;
+      add_edge(u, v, g3); ++added_edges;
+      assert(out_degree(u, g3) == 1);
+      assert(in_degree(u, g3) == 1);
+      graph_traits<Graph3>::edge_descriptor e = *out_edges(u, g3).first;
+      assert(source(e, g3) == u);
+      assert(target(e, g3) == v);
+      e = *in_edges(u, g3).first;
+      assert(source(e, g3) == v);
+      assert(target(e, g3) == u);
+      assert(out_degree(v, g3) == 1);
+      assert(in_degree(v, g3) == 1);
+      e = *out_edges(v, g3).first;
+      assert(source(e, g3) == v);
+      assert(target(e, g3) == u);
+      e = *in_edges(v, g3).first;
+      assert(source(e, g3) == u);
+      assert(target(e, g3) == v);
+
+      assert(*adjacent_vertices(u, g3).first == v);
+      assert(*adjacent_vertices(v, g3).first == u);
+      std::cout << "Processor #" << process_id(pg) << ": " << num_edges(g3)
+                << " edges.\n";
+    }
+
+    // Add some remote edges
+    for (boost::tie(v, v_end) = vertices(g3); v != v_end; ++v) {
+      graph_traits<Graph1>::vertex_descriptor other = *v;
+      other.owner = (other.owner + 1) % num_processes(pg);
+      other.local = 0;
+      add_edge(*v, other, g3); ++added_edges;
+
+      std::cout << "Adding edge from processor " << process_id(pg)
+                << " to processor " << other.owner << std::endl;
+    }
+
+    synchronize(g3);
+    assert(std::distance(edges(g3).first, edges(g3).second) == (ptrdiff_t)added_edges);
+    assert(num_edges(g3) == added_edges);
+
+    // Verify the remote edges
+    if (num_vertices(g3) >= 2) {
+      graph_traits<Graph3>::vertex_iterator vi = vertices(g3).first;
+      graph_traits<Graph3>::vertex_descriptor u = *vi++;
+
+      int prior_processor = (process_id(pg) + num_processes(pg) - 1)
+        % num_processes(pg);
+      const int n = 20;
+      graph_traits<Graph3>::vertices_size_type vertices_in_prior = (n / num_processes(pg))
+        + (n % num_processes(pg) > prior_processor? 1 : 0);
+      if (in_degree(u, g3) != vertices_in_prior + 2) {
+        std::cerr << "#" << process_id(pg) << ": " << in_degree(u, g3)
+                  << " != " << vertices_in_prior + 2 << std::endl;
+      }
+
+      assert(in_degree(u, g3) == vertices_in_prior + 2);
+      assert(out_degree(u, g3) == vertices_in_prior + 2);
+    }
+
+    local_subgraph<Graph3> local_g3(g3);
+    edges(local_g3);
+    vertices(local_g3);
+    if (num_vertices(local_g3) > 0) {
+      out_edges(*vertices(local_g3).first, local_g3);
+      in_edges(*vertices(local_g3).first, local_g3);
+      adjacent_vertices(*vertices(local_g3).first, local_g3);
+      remove_out_edge_if(*vertices(g3).first, never<Graph3>(), g3);
+      remove_in_edge_if(*vertices(g3).first, never<Graph3>(), g3);
+      if (false) {
+        // Removing an edge from two processes concurrently is not yet
+        // supported.
+        clear_vertex(*vertices(g3).first, g3);
+        remove_vertex(*vertices(g3).first, g3);
+      }
+    }
+    remove_edge_if(never<Graph3>(), g3);
+  }
+
+  return 0;
+}
diff --git a/src/boost/libs/graph_parallel/test/distributed_betweenness_centrality_test.cpp b/src/boost/libs/graph_parallel/test/distributed_betweenness_centrality_test.cpp
new file mode 100644
index 00000000..90c93143
--- /dev/null
+++ b/src/boost/libs/graph_parallel/test/distributed_betweenness_centrality_test.cpp
@@ -0,0 +1,295 @@
+// Copyright (C) 2006 The Trustees of Indiana University.
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+//  Authors: Nick Edmonds
+//           Andrew Lumsdaine
+
+#include <boost/graph/use_mpi.hpp>
+
+#define CSR
+
+#ifdef CSR
+#  include <boost/graph/distributed/compressed_sparse_row_graph.hpp>
+#else
+#  include <boost/graph/distributed/adjacency_list.hpp>
+#endif
+
+#include <boost/config.hpp>
+#include <boost/throw_exception.hpp>
+#include <boost/graph/distributed/mpi_process_group.hpp>
+#include <boost/graph/distributed/concepts.hpp>
+#include <boost/graph/erdos_renyi_generator.hpp>
+#include <boost/graph/distributed/betweenness_centrality.hpp>
+#include <boost/random/linear_congruential.hpp>
+#include <boost/graph/graphviz.hpp>
+#include <boost/property_map/vector_property_map.hpp>
+#include <boost/test/minimal.hpp>
+
+#ifdef BOOST_NO_EXCEPTIONS
+void
+boost::throw_exception(std::exception const& ex)
+{
+    std::cout << ex.what() << std::endl;
+    abort();
+}
+#endif
+
+/****************************************************************************
+ * Edge weight generator iterator                                           *
+ ****************************************************************************/
+template<typename F, typename RandomGenerator>
+class generator_iterator
+{
+public:
+  typedef std::input_iterator_tag iterator_category;
+  typedef typename F::result_type value_type;
+  typedef const value_type&       reference;
+  typedef const value_type*       pointer;
+  typedef void                    difference_type;
+
+  explicit 
+  generator_iterator(RandomGenerator& gen, const F& f = F()) 
+    : f(f), gen(&gen) 
+  { 
+    value = this->f(gen); 
+  }
+
+  reference operator*() const  { return value; }
+  pointer   operator->() const { return &value; }
+
+  generator_iterator& operator++()
+  {
+    value = f(*gen);
+    return *this;
+  }
+
+  generator_iterator operator++(int)
+  {
+    generator_iterator temp(*this);
+    ++(*this);
+    return temp;
+  }
+
+  bool operator==(const generator_iterator& other) const
+  { return f == other.f; }
+
+  bool operator!=(const generator_iterator& other) const
+  { return !(*this == other); }
+
+private:
+  F f;
+  RandomGenerator* gen;
+  value_type value;
+};
+
+template<typename F, typename RandomGenerator>
+inline generator_iterator<F, RandomGenerator> 
+make_generator_iterator( RandomGenerator& gen, const F& f)
+{ return generator_iterator<F, RandomGenerator>(gen, f); }
+
+using namespace boost;
+using boost::graph::distributed::mpi_process_group;
+
+typedef int weight_type;
+
+struct WeightedEdge {
+  WeightedEdge(weight_type weight = 0) : weight(weight) { }
+  
+  weight_type weight;
+
+  template<typename Archiver>
+  void serialize(Archiver& ar, const unsigned int /*version*/)
+  {
+    ar & weight;
+  }
+};
+
+int test_main(int argc, char* argv[])
+{
+  mpi::environment env(argc, argv);
+
+#ifdef CSR
+  typedef compressed_sparse_row_graph<directedS, no_property, WeightedEdge, 
+                                      no_property, distributedS<mpi_process_group> >
+    Graph;
+  typedef compressed_sparse_row_graph<directedS, no_property, WeightedEdge>
+    seqGraph;
+#else
+  typedef adjacency_list<vecS, 
+                         distributedS<mpi_process_group, vecS>,
+                         directedS,
+                         no_property,
+                         property<edge_weight_t, int> > Graph;
+
+  typedef adjacency_list<vecS, vecS, directedS, no_property, 
+                         property<edge_weight_t, int> > seqGraph;
+#endif
+
+
+  typedef sorted_erdos_renyi_iterator<minstd_rand, Graph> ERIter;
+
+  int n = 100;
+  double prob = 0.1;
+  int C = 3;
+
+  minstd_rand gen;
+
+  gen.seed(1); 
+
+  // NOTE: Weighted betweenness centrality only works with non-zero edge weights
+
+  // Build graphs
+  Graph g(edges_are_sorted, ERIter(gen, n, prob), ERIter(), 
+          make_generator_iterator(gen, uniform_int<int>(1, C)),
+          n);
+
+  gen.seed(1);  // Re-seed PRNG so we get the same graph
+
+  seqGraph sg(
+#ifdef CSR
+              edges_are_sorted,
+#endif
+              ERIter(gen, n, prob), ERIter(), 
+              make_generator_iterator(gen, uniform_int<int>(1, C)),
+              n);
+
+  // Test Betweenness Centrality
+  typedef property_map<Graph, vertex_index_t>::const_type IndexMap;
+  typedef iterator_property_map<std::vector<int>::iterator, IndexMap>
+    CentralityMap;
+
+  std::vector<int> centralityS(num_vertices(g), 0);
+  CentralityMap centrality(centralityS.begin(), get(vertex_index, g));
+
+  brandes_betweenness_centrality(g, centrality);
+
+  std::vector<int> weightedCentralityS(num_vertices(g), 0);
+  CentralityMap weightedCentrality(weightedCentralityS.begin(), get(vertex_index, g));
+
+  brandes_betweenness_centrality(g, centrality_map(weightedCentrality).
+#ifdef CSR
+                                 weight_map(get(&WeightedEdge::weight, g)));
+#else
+                                 weight_map(get(edge_weight, g)));
+#endif
+
+  int g_cpd = central_point_dominance(g, centrality);
+
+  //
+  //  Non-distributed (embarassingly parallel) Betweenness Centrality
+  //
+  typedef boost::graph::parallel::process_group_type<Graph>::type 
+    process_group_type;
+
+  process_group_type pg = process_group(g);
+
+  process_group_type::process_id_type id = process_id(pg);
+  process_group_type::process_size_type p = num_processes(pg);    
+
+  typedef property_map<seqGraph, vertex_index_t>::const_type seqIndexMap;
+  typedef iterator_property_map<std::vector<int>::iterator, seqIndexMap> seqCentralityMap;
+
+  std::vector<int> nonDistributedCentralityS(num_vertices(sg), 0);
+  seqCentralityMap nonDistributedCentrality(nonDistributedCentralityS.begin(), get(vertex_index, sg));
+
+  std::vector<int> nonDistributedWeightedCentralityS(num_vertices(sg), 0);
+  seqCentralityMap nonDistributedWeightedCentrality(nonDistributedWeightedCentralityS.begin(), 
+                                                    get(vertex_index, sg));
+
+  non_distributed_brandes_betweenness_centrality(pg, sg, nonDistributedCentrality);
+
+  non_distributed_brandes_betweenness_centrality(pg, sg, 
+                                                 centrality_map(nonDistributedWeightedCentrality).
+#ifdef CSR
+                                                 weight_map(get(&WeightedEdge::weight, sg)));
+#else
+                                                 weight_map(get(edge_weight, sg)));
+#endif
+
+  //
+  // Verify
+  //
+  std::vector<int> seqCentralityS(num_vertices(sg), 0);
+  seqCentralityMap seqCentrality(seqCentralityS.begin(), get(vertex_index, sg));
+
+  std::vector<int> seqWeightedCentralityS(num_vertices(sg), 0);
+  seqCentralityMap seqWeightedCentrality(seqWeightedCentralityS.begin(), get(vertex_index, sg));
+
+  brandes_betweenness_centrality(sg, seqCentrality);
+
+  brandes_betweenness_centrality(sg, centrality_map(seqWeightedCentrality).
+#ifdef CSR
+                                 weight_map(get(&WeightedEdge::weight, sg)));
+#else
+                                 weight_map(get(edge_weight, sg)));
+#endif
+
+  int sg_cpd = central_point_dominance(sg, seqCentrality);
+
+  // Verify exact betweenness centrality
+  //
+  // We're cheating when we map vertices in g to vertices in sg
+  // since we know that g is using the block distribution here
+  typedef property_map<Graph, vertex_owner_t>::const_type OwnerMap;
+  typedef property_map<Graph, vertex_local_t>::const_type LocalMap;
+
+  OwnerMap owner = get(vertex_owner, g);
+  LocalMap local = get(vertex_local, g);
+
+  bool passed = true;
+  
+  {
+    typedef graph_traits<Graph>::vertex_iterator vertex_iterator;
+    vertex_iterator v, v_end;
+    
+    for (boost::tie(v, v_end) = vertices(g); v != v_end; ++v) {
+      if (get(centrality, *v) != seqCentralityS[(n/p) * get(owner, *v) + get(local, *v)]) {
+        std::cerr << "  " << id << ": Error - centrality of " << get(local, *v) << "@" << get(owner, *v)
+                  << " does not match the sequential result (" << get(centrality, *v) << " vs. " 
+                  << seqCentralityS[(n/p) * get(owner, *v) + get(local, *v)] << ")\n";
+        passed = false;
+      }
+      
+      if (get(weightedCentrality, *v) != seqWeightedCentralityS[(n/p) * get(owner, *v) + get(local, *v)]) {
+        std::cerr << "  " << id << ": Error - weighted centrality of " << get(local, *v) << "@" << get(owner, *v)
+                  << " does not match the sequential result (" << get(weightedCentrality, *v) << " vs. " 
+                  << seqWeightedCentralityS[(n/p) * get(owner, *v) + get(local, *v)] << ")\n";
+        passed = false;
+      }
+    }
+  }
+
+  if (id == 0) {
+    typedef graph_traits<seqGraph>::vertex_iterator vertex_iterator;
+    vertex_iterator v, v_end;
+  
+    for (boost::tie(v, v_end) = vertices(sg); v != v_end; ++v) {
+      if (get(seqCentrality, *v) != get(nonDistributedCentrality, *v)) {
+        std::cerr << "  " << id << ": Error - non-distributed centrality of " << *v
+                  << " does not match the sequential result (" << get(nonDistributedCentrality, *v) 
+                  << " vs. " << get(seqCentrality, *v) << ")\n";
+        passed = false;
+      }
+
+      if (get(seqWeightedCentrality, *v) != get(nonDistributedWeightedCentrality, *v)) {
+        std::cerr << "  " << id << ": Error - non-distributed weighted centrality of " << *v
+                  << " does not match the sequential result (" << get(nonDistributedWeightedCentrality, *v) 
+                  << " vs. " << get(seqCentrality, *v) << ")\n";
+        passed = false;
+      }
+    }
+  }
+
+  if (g_cpd != sg_cpd) {
+    passed = false;
+    std::cerr << "Central point dominance did not match the sequential result\n";
+  }
+
+  if (!passed)
+    MPI_Abort(MPI_COMM_WORLD, -1);
+
+  return 0;
+}
diff --git a/src/boost/libs/graph_parallel/test/distributed_connected_components_test.cpp b/src/boost/libs/graph_parallel/test/distributed_connected_components_test.cpp
new file mode 100644
index 00000000..e630bbb0
--- /dev/null
+++ b/src/boost/libs/graph_parallel/test/distributed_connected_components_test.cpp
@@ -0,0 +1,204 @@
+// Copyright (C) 2004-2006 The Trustees of Indiana University.
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+//  Authors: Douglas Gregor
+//           Andrew Lumsdaine
+
+#include <boost/graph/use_mpi.hpp>
+#include <boost/config.hpp>
+#include <boost/throw_exception.hpp>
+#include <boost/serialization/vector.hpp>
+#include <boost/graph/distributed/adjacency_list.hpp>
+#include <boost/graph/connected_components.hpp>
+#include <boost/graph/distributed/connected_components_parallel_search.hpp>
+#include <boost/graph/random.hpp>
+#include <boost/property_map/parallel/distributed_property_map.hpp>
+#include <boost/graph/distributed/mpi_process_group.hpp>
+#include <boost/graph/parallel/distribution.hpp>
+#include <boost/graph/erdos_renyi_generator.hpp>
+#include <boost/graph/distributed/graphviz.hpp>
+#include <iostream>
+#include <cstdlib>
+#include <iomanip>
+#include <boost/random.hpp>
+#include <boost/test/minimal.hpp>
+
+#include <boost/graph/distributed/compressed_sparse_row_graph.hpp>
+#include <boost/graph/rmat_graph_generator.hpp>
+
+#ifdef BOOST_NO_EXCEPTIONS
+void
+boost::throw_exception(std::exception const& ex)
+{
+    std::cout << ex.what() << std::endl;
+    abort();
+}
+#endif
+
+using namespace boost;
+using boost::graph::distributed::mpi_process_group;
+
+typedef double time_type;
+
+inline time_type get_time()
+{
+        return MPI_Wtime();
+}
+
+std::string print_time(time_type t)
+{
+  std::ostringstream out;
+  out << std::setiosflags(std::ios::fixed) << std::setprecision(2) << t;
+  return out.str();
+}
+
+template<typename T>
+class map_lt
+{
+public:
+  bool operator()() const { return false; }
+  bool operator()(T x, T y) const { return (owner(x) < owner(y) || (owner(x) == owner(y) && local(x) < local(y))); }
+};
+
+void 
+test_distributed_connected_components(int n, double _p, bool verify, bool emit_dot_file, int seed, bool parallel_search)
+{
+//   typedef adjacency_list<listS, 
+//                          distributedS<mpi_process_group, vecS>,
+//                          undirectedS> Graph;
+
+  typedef compressed_sparse_row_graph<directedS, no_property, no_property, no_property,
+                                      distributedS<mpi_process_group> > Graph;
+
+  minstd_rand gen;
+
+  gen.seed(seed);
+
+  mpi_process_group pg;
+  parallel::variant_distribution<mpi_process_group> distrib 
+    = parallel::block(pg, n);
+
+  minstd_rand dist_gen;
+#if 0
+  if (false) {
+    distrib = parallel::random_distribution(pg, dist_gen, n);
+  } else if (true) {
+    distrib = parallel::oned_block_cyclic(pg, 13);
+  }
+#endif
+
+//   Graph g(erdos_renyi_iterator<minstd_rand, Graph>(gen, n, _p/2),
+//           erdos_renyi_iterator<minstd_rand, Graph>(),
+//           n, pg, distrib);
+
+  int m = int(n * n * _p/2);
+  double a = 0.57, b = 0.19, c = 0.19, d = 0.05;
+
+  // Last boolean parameter makes R-MAT bidirectional
+  Graph g(sorted_unique_rmat_iterator<minstd_rand, Graph>(gen, n, m, a, b, c, d, true),
+          sorted_unique_rmat_iterator<minstd_rand, Graph>(),
+          n, pg, distrib);
+
+  synchronize(g);
+
+  std::vector<int> local_components_vec(num_vertices(g));
+  typedef iterator_property_map<std::vector<int>::iterator, property_map<Graph, vertex_index_t>::type> ComponentMap;
+  ComponentMap component(local_components_vec.begin(), get(vertex_index, g));
+
+  int num_components = 0;
+
+  time_type start = get_time();
+  if (parallel_search) {
+    num_components = connected_components_ps(g, component);
+  } else {
+    num_components = connected_components(g, component);
+  }
+  time_type end = get_time();
+  if (process_id(g.process_group()) == 0) 
+    std::cerr << "Connected Components time = " << print_time(end - start) << " seconds.\n"
+              << num_components << " components identified\n";
+
+  if ( verify )
+    {
+      if ( process_id(g.process_group()) == 0 )
+        {
+          component.set_max_ghost_cells(0);
+          for (int i = 0; i < n; ++i)
+            get(component, vertex(i, g));
+          synchronize(component);
+
+          // Check against the sequential version
+          typedef adjacency_list<listS, 
+            vecS,
+            undirectedS,
+            // Vertex properties
+            no_property,
+            // Edge properties
+            no_property > Graph2;
+          
+          gen.seed(seed);
+          
+//        Graph2 g2(erdos_renyi_iterator<minstd_rand, Graph>(gen, n, _p/2),
+//                  erdos_renyi_iterator<minstd_rand, Graph>(),
+//                  n);
+
+          Graph2 g2( sorted_unique_rmat_iterator<minstd_rand, Graph>(gen, n, m, a, b, c, d, true),
+                     sorted_unique_rmat_iterator<minstd_rand, Graph>(),
+                     n);
+          
+          std::vector<int> component2 (n);
+          int tmp;
+          tmp = connected_components(g2, make_iterator_property_map(component2.begin(), get(vertex_index, g2)));
+          std::cerr << "Verifier found " << tmp << " components" << std::endl;
+          
+          // Make sure components and component2 match
+          std::map<int, int> c2c;
+          int i;
+          // This fails if there are more components in 'component' than
+          // 'component2' because multiple components in 'component' may 
+          // legitimately map to the same component number in 'component2'.
+          // We can either test the mapping in the opposite direction or
+          // just assert that the numbers of components found by both 
+          // algorithms is the same
+          for ( i = 0; i < n; i++ )
+            if ( c2c.find( get(component, vertex(i, g)) ) == c2c.end() )
+              c2c[get(component, vertex(i, g))] = component2[i];
+            else
+              if ( c2c[get(component, vertex(i, g))] != component2[i] )
+                break;
+          
+          if ( i < n || num_components != tmp) {
+            printf("Unable to verify CC result...\n");
+          } else
+            printf("Passed verification... %i connected components\n", 
+                   (int)c2c.size());
+        }
+      else 
+        {
+          synchronize(component);
+        }
+      if ( emit_dot_file )
+        write_graphviz("cc.dot", g, paint_by_number(component));
+    }
+}
+
+int test_main(int argc, char* argv[])
+{
+  mpi::environment env(argc, argv);
+
+  if ( argc < 6 ) {
+    test_distributed_connected_components(10000, 0.001, true, false, 1, false);
+    test_distributed_connected_components(10000, 0.001, true, false, 1, true);
+  }
+  else
+    test_distributed_connected_components
+      (atoi(argv[1]), atof(argv[2]), 
+       argv[3]==std::string("true"), argv[4]==std::string("true"),
+       argc == 6? 1 : atoi(argv[6]),
+       argv[5]==std::string("true"));
+
+  return 0;
+}
diff --git a/src/boost/libs/graph_parallel/test/distributed_csr_algorithm_test.cpp b/src/boost/libs/graph_parallel/test/distributed_csr_algorithm_test.cpp
new file mode 100644
index 00000000..0653cc91
--- /dev/null
+++ b/src/boost/libs/graph_parallel/test/distributed_csr_algorithm_test.cpp
@@ -0,0 +1,369 @@
+// Copyright (C) 2006 The Trustees of Indiana University.
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+//  Authors: Nick Edmonds
+//           Andrew Lumsdaine
+
+// A test of the distributed compressed sparse row graph type
+#include <boost/graph/use_mpi.hpp>
+#include <boost/config.hpp>
+#include <boost/throw_exception.hpp>
+#include <boost/graph/distributed/compressed_sparse_row_graph.hpp>
+#include <boost/graph/distributed/mpi_process_group.hpp>
+#include <boost/graph/distributed/concepts.hpp>
+
+#include <boost/graph/erdos_renyi_generator.hpp>
+#include <boost/graph/small_world_generator.hpp>
+#include <boost/graph/rmat_graph_generator.hpp>
+
+#include <boost/graph/breadth_first_search.hpp>
+#include <boost/graph/depth_first_search.hpp>
+#include <boost/graph/distributed/delta_stepping_shortest_paths.hpp>
+#include <boost/graph/dijkstra_shortest_paths.hpp>
+#include <boost/graph/distributed/page_rank.hpp>
+#include <boost/graph/distributed/boman_et_al_graph_coloring.hpp>
+#include <boost/graph/connected_components.hpp>
+#include <boost/graph/strong_components.hpp>
+#include <boost/graph/distributed/betweenness_centrality.hpp>
+#include <boost/graph/distributed/dehne_gotz_min_spanning_tree.hpp>
+#include <boost/graph/distributed/st_connected.hpp>
+
+#if 0 // Contains internal AdjList types not present in CSR graph
+#  include <boost/graph/distributed/connected_components_parallel_search.hpp>
+#endif
+
+#include <boost/graph/distributed/vertex_list_adaptor.hpp> // Needed for MST
+
+#include <boost/random/linear_congruential.hpp>
+#include <boost/graph/graphviz.hpp>
+#include <boost/property_map/vector_property_map.hpp>
+#include <boost/test/minimal.hpp>
+
+#ifdef BOOST_NO_EXCEPTIONS
+void
+boost::throw_exception(std::exception const& ex)
+{
+    std::cout << ex.what() << std::endl;
+    abort();
+}
+#endif
+
+
+/****************************************************************************
+ * Edge weight generator iterator                                           *
+ ****************************************************************************/
+template<typename F, typename RandomGenerator>
+class generator_iterator
+{
+public:
+  typedef std::input_iterator_tag iterator_category;
+  typedef typename F::result_type value_type;
+  typedef const value_type&       reference;
+  typedef const value_type*       pointer;
+  typedef void                    difference_type;
+
+  explicit 
+  generator_iterator(RandomGenerator& gen, const F& f = F()) 
+    : f(f), gen(&gen) 
+  { 
+    value = this->f(gen); 
+  }
+
+  reference operator*() const  { return value; }
+  pointer   operator->() const { return &value; }
+
+  generator_iterator& operator++()
+  {
+    value = f(*gen);
+    return *this;
+  }
+
+  generator_iterator operator++(int)
+  {
+    generator_iterator temp(*this);
+    ++(*this);
+    return temp;
+  }
+
+  bool operator==(const generator_iterator& other) const
+  { return f == other.f; }
+
+  bool operator!=(const generator_iterator& other) const
+  { return !(*this == other); }
+
+private:
+  F f;
+  RandomGenerator* gen;
+  value_type value;
+};
+
+template<typename F, typename RandomGenerator>
+inline generator_iterator<F, RandomGenerator> 
+make_generator_iterator( RandomGenerator& gen, const F& f)
+{ return generator_iterator<F, RandomGenerator>(gen, f); }
+
+
+/****************************************************************************
+ * Printing DFS Visitor                                                     *
+ ****************************************************************************/
+
+struct printing_dfs_visitor
+{
+  template<typename Vertex, typename Graph>
+  void initialize_vertex(Vertex v, const Graph& g)
+  {
+    vertex_event("initialize_vertex", v, g);
+  }
+
+  template<typename Vertex, typename Graph>
+  void start_vertex(Vertex v, const Graph& g)
+  {
+    vertex_event("start_vertex", v, g);
+  }
+
+  template<typename Vertex, typename Graph>
+  void discover_vertex(Vertex v, const Graph& g)
+  {
+    vertex_event("discover_vertex", v, g);
+  }
+
+  template<typename Edge, typename Graph>
+  void examine_edge(Edge e, const Graph& g)
+  {
+    edge_event("examine_edge", e, g);
+  }
+
+  template<typename Edge, typename Graph>
+  void tree_edge(Edge e, const Graph& g)
+  {
+    edge_event("tree_edge", e, g);
+  }
+
+  template<typename Edge, typename Graph>
+  void back_edge(Edge e, const Graph& g)
+  {
+    edge_event("back_edge", e, g);
+  }
+
+  template<typename Edge, typename Graph>
+  void forward_or_cross_edge(Edge e, const Graph& g)
+  {
+    edge_event("forward_or_cross_edge", e, g);
+  }
+
+  template<typename Vertex, typename Graph>
+  void finish_vertex(Vertex v, const Graph& g)
+  {
+    vertex_event("finish_vertex", v, g);
+  }
+
+private:
+  template<typename Vertex, typename Graph>
+  void vertex_event(const char* name, Vertex v, const Graph& g)
+  {
+    std::cerr << "#" << process_id(g.process_group()) << ": " << name << "("
+              << get_vertex_name(v, g) << ": " << local(v) << "@" << owner(v)
+              << ")\n";
+  }
+
+  template<typename Edge, typename Graph>
+  void edge_event(const char* name, Edge e, const Graph& g)
+  {
+    std::cerr << "#" << process_id(g.process_group()) << ": " << name << "("
+              << get_vertex_name(source(e, g), g) << ": "
+              << local(source(e, g)) << "@" << owner(source(e, g)) << ", "
+              << get_vertex_name(target(e, g), g) << ": "
+              << local(target(e, g)) << "@" << owner(target(e, g)) << ")\n";
+  }
+
+};
+
+using namespace boost;
+using boost::graph::distributed::mpi_process_group;
+
+typedef int weight_type;
+
+struct WeightedEdge {
+  WeightedEdge(weight_type weight = 0) : weight(weight) { }
+  
+  weight_type weight;
+
+  template<typename Archiver>
+  void serialize(Archiver& ar, const unsigned int /*version*/)
+  {
+    ar & weight;
+  }
+};
+
+struct VertexProperties {
+  VertexProperties(int d = 0)
+    : distance(d) { }
+
+  int distance;
+
+  template<typename Archiver>
+  void serialize(Archiver& ar, const unsigned int /*version*/)
+  {
+    ar & distance;
+  }
+};
+
+int test_main(int argc, char* argv[])
+{
+  mpi::environment env(argc, argv);
+
+  typedef compressed_sparse_row_graph<directedS, VertexProperties, WeightedEdge, 
+                                      no_property, distributedS<mpi_process_group> >
+    Digraph;
+
+  // Make sure we can build graphs using common graph generators
+  typedef sorted_erdos_renyi_iterator<minstd_rand, Digraph> ERIter;
+  typedef small_world_iterator<minstd_rand, Digraph> SWIter;
+  typedef sorted_rmat_iterator<minstd_rand, Digraph> RMATIter;
+
+  typedef graph_traits<Digraph>::edge_descriptor edge_descriptor;
+
+  int n = 40;
+  int k = 3;
+  double prob = 0.1;
+  int C = 10;
+  double a = 0.5, b = 0.1, c = 0.25, d = 0.15;
+  int iterations = 50;
+  int num_colors = n / 10;
+  int lookahead = C / 10;
+
+  minstd_rand gen;
+
+  // Directed Graphs
+  Digraph g(ERIter(gen, n, prob), ERIter(), 
+            make_generator_iterator(gen, uniform_int<int>(0, C)),
+            n);
+  Digraph g2(SWIter(gen, n, k, prob), SWIter(), n);
+  Digraph g3(RMATIter(gen, n, size_t(n*n*prob), a, b, c, d), RMATIter(), n);
+
+  // Test BFS
+  breadth_first_search(g, vertex(0, g), visitor(bfs_visitor<>()));
+
+  // Test SSSP Algorithms
+  graph::distributed::delta_stepping_shortest_paths(g, 
+                                                    vertex(0, g),
+                                                    dummy_property_map(),
+                                                    get(&VertexProperties::distance, g),
+                                                    get(&WeightedEdge::weight, g));
+
+  dijkstra_shortest_paths(g, vertex(0, g),
+                          distance_map(get(&VertexProperties::distance, g)).
+                          weight_map(get(&WeightedEdge::weight, g)));
+
+  dijkstra_shortest_paths(g, vertex(0, g),
+                          distance_map(get(&VertexProperties::distance, g)).
+                          weight_map(get(&WeightedEdge::weight, g)).
+                          lookahead(lookahead));
+
+  // Test PageRank
+  using boost::graph::n_iterations;
+
+  std::vector<double> ranks(num_vertices(g));
+
+  page_rank(g,
+            make_iterator_property_map(ranks.begin(),
+                                       get(boost::vertex_index, g)),
+            n_iterations(iterations), 0.85, vertex(0, g));
+
+  // Test Graph Coloring
+  typedef property_map<Digraph, vertex_index_t>::type vertex_index_map;
+
+  std::vector<int> colors_vec(num_vertices(g));
+  iterator_property_map<int*, vertex_index_map> color(&colors_vec[0],
+                                                      get(vertex_index, g));
+
+  graph::boman_et_al_graph_coloring(g, color, num_colors);
+
+  // Test DFS
+  //
+  // DFS requires an undirected graph, currently CSR graphs must be directed
+#if 0
+  std::vector<vertex_descriptor> parent(num_vertices(g));
+  std::vector<vertex_descriptor> explore(num_vertices(g));
+
+  boost::graph::tsin_depth_first_visit
+    (g,
+     vertex(0, g),
+     printing_dfs_visitor(),
+     color,
+     make_iterator_property_map(parent.begin(), get(vertex_index, g)),
+     make_iterator_property_map(explore.begin(), get(vertex_index, g)),
+     get(vertex_index, g));
+#endif
+
+  // Test S-T Connected
+  st_connected(g, vertex(0, g), vertex(1, g), color, get(vertex_owner, g));
+
+  // Test Connected Components
+  //
+  // CC requires an undirected graph, currently CSR graphs must be directed
+#if 0
+  std::vector<int> local_components_vec(num_vertices(g));
+  typedef iterator_property_map<std::vector<int>::iterator, 
+                                vertex_index_map> ComponentMap;
+  ComponentMap component(local_components_vec.begin(), get(vertex_index, g));
+
+  assert(connected_components(g, component) == 
+         connected_components_ps(g, component));
+#endif
+
+  // Test Betweenness Centrality
+  //
+  // Betweenness Centrality is broken at the moment
+  typedef iterator_property_map<std::vector<int>::iterator, vertex_index_map> 
+    CentralityMap;
+
+  std::vector<int> centralityS(num_vertices(g), 0);
+  CentralityMap centrality(centralityS.begin(), get(vertex_index, g));
+
+  brandes_betweenness_centrality(g, centrality);
+
+  //
+  // Test MST
+  //
+  std::vector<edge_descriptor> mst_edges;
+
+  dense_boruvka_minimum_spanning_tree(make_vertex_list_adaptor(g), 
+                                      get(&WeightedEdge::weight, g),
+                                      std::back_inserter(mst_edges));
+
+  mst_edges.clear();
+  merge_local_minimum_spanning_trees(make_vertex_list_adaptor(g), 
+                                     get(&WeightedEdge::weight, g),
+                                     std::back_inserter(mst_edges));
+
+  mst_edges.clear();
+  boruvka_then_merge(make_vertex_list_adaptor(g), 
+                     get(&WeightedEdge::weight, g),
+                     std::back_inserter(mst_edges));
+
+  mst_edges.clear();
+  boruvka_mixed_merge(make_vertex_list_adaptor(g), 
+                      get(&WeightedEdge::weight, g),
+                      std::back_inserter(mst_edges));
+
+  // Test Strong Components
+  //
+  // Can't build reverse graph of a CSR graph
+#if 0
+  std::vector<int> local_components_vec(num_vertices(g));
+  typedef iterator_property_map<std::vector<int>::iterator, 
+                                vertex_index_map> ComponentMap;
+  ComponentMap component(local_components_vec.begin(), get(vertex_index, g));
+
+  int num_components = strong_components(g, component);
+#endif
+
+  std::ofstream out("dcsr.dot");
+  write_graphviz(out, g);
+
+  return 0;
+}
diff --git a/src/boost/libs/graph_parallel/test/distributed_csr_test.cpp b/src/boost/libs/graph_parallel/test/distributed_csr_test.cpp
new file mode 100644
index 00000000..a6833ecb
--- /dev/null
+++ b/src/boost/libs/graph_parallel/test/distributed_csr_test.cpp
@@ -0,0 +1,103 @@
+// Copyright (C) 2006 The Trustees of Indiana University.
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+//  Authors: Douglas Gregor
+//           Andrew Lumsdaine
+
+// A test of the distributed compressed sparse row graph type
+#include <boost/graph/use_mpi.hpp>
+#include <boost/config.hpp>
+#include <boost/throw_exception.hpp>
+#include <boost/serialization/vector.hpp>
+#include <boost/graph/distributed/compressed_sparse_row_graph.hpp>
+#include <boost/graph/distributed/mpi_process_group.hpp>
+#include <boost/graph/distributed/concepts.hpp>
+#include <boost/graph/erdos_renyi_generator.hpp>
+#include <boost/random/linear_congruential.hpp>
+#include <boost/graph/breadth_first_search.hpp>
+#include <boost/graph/graphviz.hpp>
+#include <boost/property_map/vector_property_map.hpp>
+#include <boost/test/minimal.hpp>
+
+#ifdef BOOST_NO_EXCEPTIONS
+void
+boost::throw_exception(std::exception const& ex)
+{
+    std::cout << ex.what() << std::endl;
+    abort();
+}
+#endif
+
+using namespace boost;
+using boost::graph::distributed::mpi_process_group;
+
+void concept_checks()
+{
+  typedef compressed_sparse_row_graph<directedS, no_property, no_property, no_property, 
+                                      distributedS<mpi_process_group> >
+    Digraph;
+  typedef graph_traits<Digraph>::vertex_descriptor vertex_descriptor;
+  typedef graph_traits<Digraph>::edge_descriptor edge_descriptor;
+
+  function_requires< GraphConcept<Digraph> >();
+  function_requires< IncidenceGraphConcept<Digraph> >();
+  function_requires< AdjacencyGraphConcept<Digraph> >();
+
+  function_requires< DistributedVertexListGraphConcept<Digraph> >();
+  function_requires< DistributedEdgeListGraphConcept<Digraph> >();
+
+  function_requires<
+    ReadablePropertyGraphConcept<Digraph, vertex_descriptor, vertex_global_t> 
+    >();
+  function_requires<
+    ReadablePropertyGraphConcept<Digraph, vertex_descriptor, vertex_owner_t>
+    >();
+  function_requires<
+    ReadablePropertyGraphConcept<Digraph, vertex_descriptor, vertex_local_t>
+    >();
+  function_requires<
+    ReadablePropertyGraphConcept<Digraph, vertex_descriptor, vertex_index_t>
+    >();
+
+  function_requires<
+    ReadablePropertyGraphConcept<Digraph, edge_descriptor, edge_global_t>
+    >();
+
+  // DPG TBD: edge_owner, edge_local property maps
+
+  function_requires<
+    ReadablePropertyGraphConcept<Digraph, edge_descriptor, edge_index_t>
+    >();
+
+  // Check default construction
+  Digraph g;
+}
+
+int test_main(int argc, char* argv[])
+{
+  mpi::environment env(argc, argv);
+
+  concept_checks();
+
+  typedef compressed_sparse_row_graph<directedS, no_property, no_property, no_property, 
+                                      distributedS<mpi_process_group> >
+    Digraph;
+
+  // Build an Erdos-Renyi graph to test with
+  typedef sorted_erdos_renyi_iterator<minstd_rand, Digraph> ERIter;
+
+  int n = 40;
+  double prob = 0.1;
+
+  minstd_rand gen;
+  Digraph g(ERIter(gen, n, prob), ERIter(), n);
+
+  breadth_first_search(g, vertex(0, g), visitor(bfs_visitor<>()));
+
+  std::ofstream out("dcsr.dot");
+  write_graphviz(out, g);
+  return 0;
+}
diff --git a/src/boost/libs/graph_parallel/test/distributed_dfs_test.cpp b/src/boost/libs/graph_parallel/test/distributed_dfs_test.cpp
new file mode 100644
index 00000000..d21491b3
--- /dev/null
+++ b/src/boost/libs/graph_parallel/test/distributed_dfs_test.cpp
@@ -0,0 +1,167 @@
+//=======================================================================
+// Copyright 2001 Jeremy G. Siek, Andrew Lumsdaine, Lie-Quan Lee,
+// Copyright 2004 Douglas Gregor
+// Copyright (C) 2006-2008 
+
+// Distributed under the Boost Software License, Version 1.0. (See
+// accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+//=======================================================================
+#include <boost/graph/use_mpi.hpp>
+#include <boost/config.hpp>
+#include <boost/throw_exception.hpp>
+#include <boost/serialization/vector.hpp>
+#include <boost/graph/depth_first_search.hpp>
+#include <boost/graph/distributed/mpi_process_group.hpp>
+#include <boost/graph/distributed/adjacency_list.hpp>
+#include <boost/test/minimal.hpp>
+#include <iostream>
+
+#ifdef BOOST_NO_EXCEPTIONS
+void
+boost::throw_exception(std::exception const& ex)
+{
+    std::cout << ex.what() << std::endl;
+    abort();
+}
+#endif
+
+using namespace boost;
+using boost::graph::distributed::mpi_process_group;
+
+// Set up the vertex names
+enum vertex_id_t { u, v, w, x, y, z, N };
+char vertex_names[] = { 'u', 'v', 'w', 'x', 'y', 'z' };
+
+template<typename Vertex, typename Graph>
+char get_vertex_name(Vertex v,  const Graph& g)
+{
+  return vertex_names[g.distribution().global(owner(v), local(v))];
+}
+
+struct printing_dfs_visitor
+{
+  template<typename Vertex, typename Graph>
+  void initialize_vertex(Vertex v, const Graph& g)
+  {
+    vertex_event("initialize_vertex", v, g);
+  }
+
+  template<typename Vertex, typename Graph>
+  void start_vertex(Vertex v, const Graph& g)
+  {
+    vertex_event("start_vertex", v, g);
+  }
+
+  template<typename Vertex, typename Graph>
+  void discover_vertex(Vertex v, const Graph& g)
+  {
+    vertex_event("discover_vertex", v, g);
+  }
+
+  template<typename Edge, typename Graph>
+  void examine_edge(Edge e, const Graph& g)
+  {
+    edge_event("examine_edge", e, g);
+  }
+
+  template<typename Edge, typename Graph>
+  void tree_edge(Edge e, const Graph& g)
+  {
+    edge_event("tree_edge", e, g);
+  }
+
+  template<typename Edge, typename Graph>
+  void back_edge(Edge e, const Graph& g)
+  {
+    edge_event("back_edge", e, g);
+  }
+
+  template<typename Edge, typename Graph>
+  void forward_or_cross_edge(Edge e, const Graph& g)
+  {
+    edge_event("forward_or_cross_edge", e, g);
+  }
+
+  template<typename Vertex, typename Graph>
+  void finish_vertex(Vertex v, const Graph& g)
+  {
+    vertex_event("finish_vertex", v, g);
+  }
+
+private:
+  template<typename Vertex, typename Graph>
+  void vertex_event(const char* name, Vertex v, const Graph& g)
+  {
+    std::cerr << "#" << process_id(g.process_group()) << ": " << name << "("
+              << get_vertex_name(v, g) << ": " << local(v) << "@" << owner(v)
+              << ")\n";
+  }
+
+  template<typename Edge, typename Graph>
+  void edge_event(const char* name, Edge e, const Graph& g)
+  {
+    std::cerr << "#" << process_id(g.process_group()) << ": " << name << "("
+              << get_vertex_name(source(e, g), g) << ": "
+              << local(source(e, g)) << "@" << owner(source(e, g)) << ", "
+              << get_vertex_name(target(e, g), g) << ": "
+              << local(target(e, g)) << "@" << owner(target(e, g)) << ")\n";
+  }
+
+};
+
+void
+test_distributed_dfs()
+{
+  typedef adjacency_list<listS,
+                         distributedS<mpi_process_group, vecS>,
+                         undirectedS,
+                         // Vertex properties
+                         property<vertex_color_t, default_color_type> >
+    Graph;
+  typedef graph_traits<Graph>::vertex_descriptor vertex_descriptor;
+
+  // Specify the edges in the graph
+  typedef std::pair<int, int> E;
+  E edge_array[] = { E(u, v), E(u, w), E(u, x), E(x, v), E(y, x),
+                     E(v, y), E(w, y), E(w, z), E(z, z) };
+  Graph g(edge_array, edge_array + sizeof(edge_array) / sizeof(E), N);
+
+  std::vector<vertex_descriptor> parent(num_vertices(g));
+  std::vector<vertex_descriptor> explore(num_vertices(g));
+
+  boost::graph::tsin_depth_first_visit
+    (g,
+     vertex(u, g),
+     printing_dfs_visitor(),
+     get(vertex_color, g),
+     make_iterator_property_map(parent.begin(), get(vertex_index, g)),
+     make_iterator_property_map(explore.begin(), get(vertex_index, g)),
+     get(vertex_index, g));
+
+#if 0
+  std::size_t correct_parents1[N] = {u, u, y, y, v, w};
+  std::size_t correct_parents2[N] = {u, u, y, v, x, w};
+#endif
+
+  for (std::size_t i = 0; i < N; ++i) {
+    vertex_descriptor v = vertex(i, g);
+    if (owner(v) == process_id(g.process_group())) {
+      std::cout  << "parent(" << get_vertex_name(v, g) << ") = "
+                 << get_vertex_name(parent[v.local], g) << std::endl;
+
+    }
+  }
+
+  if (false) {
+    depth_first_visit(g, vertex(u, g), printing_dfs_visitor());
+  }
+}
+
+int
+test_main(int argc, char* argv[])
+{
+  mpi::environment env(argc, argv);
+  test_distributed_dfs();
+  return 0;
+}
diff --git a/src/boost/libs/graph_parallel/test/distributed_dimacs_reader.cpp b/src/boost/libs/graph_parallel/test/distributed_dimacs_reader.cpp
new file mode 100644
index 00000000..a2e3cb7b
--- /dev/null
+++ b/src/boost/libs/graph_parallel/test/distributed_dimacs_reader.cpp
@@ -0,0 +1,85 @@
+// Copyright (C) 2006 The Trustees of Indiana University.
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+//  Authors: Brian Barrett
+//           Nick Edmonds
+//           Andrew Lumsdaine
+
+#include <boost/graph/use_mpi.hpp>
+#include <boost/config.hpp>
+#include <boost/throw_exception.hpp>
+#include <boost/graph/distributed/adjacency_list.hpp>
+#include <boost/property_map/parallel/distributed_property_map.hpp>
+#include <boost/graph/distributed/mpi_process_group.hpp>
+#include <boost/graph/dimacs.hpp>
+#include <boost/graph/graphviz.hpp>
+#include <boost/test/minimal.hpp>
+
+#include <iostream>
+#include <cstdlib>
+#include <iomanip>
+#include <fstream>
+
+#ifdef BOOST_NO_EXCEPTIONS
+void
+boost::throw_exception(std::exception const& ex)
+{
+  std::cout << ex.what() << std::endl;
+  abort();
+}
+#endif
+
+using namespace boost;
+using namespace boost::graph;
+using boost::graph::distributed::mpi_process_group;
+
+typedef double time_type;
+
+inline time_type get_time()
+{
+  return MPI_Wtime();
+}
+
+std::string print_time(time_type t)
+{
+  std::ostringstream out;
+  out << std::setiosflags(std::ios::fixed) << std::setprecision(2) << t;
+  return out.str();
+}
+
+void
+test_dimacs_reader(const char *filename)
+{
+  mpi_process_group pg;
+
+  typedef adjacency_list<vecS, 
+        distributedS<mpi_process_group, vecS>,
+        undirectedS> Graph;
+
+  std::ifstream file(filename);
+  dimacs_basic_reader reader = dimacs_basic_reader(file, false);
+  dimacs_basic_reader end;
+  boost::parallel::variant_distribution<mpi_process_group> distrib = 
+      boost::parallel::block(pg, reader.n_vertices());
+
+  Graph g(dimacs_edge_iterator<dimacs_basic_reader>(reader), 
+          dimacs_edge_iterator<dimacs_basic_reader>(end),
+          reader.n_vertices(), pg, distrib);;
+
+    // write_graphviz("reader.dot", g);
+}
+
+int
+test_main(int argc, char* argv[])
+{
+  mpi::environment env(argc, argv);
+
+  if (argc == 2) { 
+      test_dimacs_reader(argv[1]);
+  }
+
+  return 0;
+}
diff --git a/src/boost/libs/graph_parallel/test/distributed_graph_coloring_test.cpp b/src/boost/libs/graph_parallel/test/distributed_graph_coloring_test.cpp
new file mode 100644
index 00000000..2b219c0a
--- /dev/null
+++ b/src/boost/libs/graph_parallel/test/distributed_graph_coloring_test.cpp
@@ -0,0 +1,101 @@
+// Copyright (C) 2005, 2006 The Trustees of Indiana University.
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+//  Authors: Douglas Gregor
+//           Andrew Lumsdaine
+#define PBGL_ACCOUNTING
+
+#include <boost/graph/use_mpi.hpp>
+#include <boost/config.hpp>
+#include <boost/throw_exception.hpp>
+#include <boost/serialization/vector.hpp>
+#include <boost/graph/distributed/boman_et_al_graph_coloring.hpp>
+#include <boost/graph/distributed/mpi_process_group.hpp>
+#include <boost/lexical_cast.hpp>
+#include <boost/graph/parallel/distribution.hpp>
+#include <boost/graph/erdos_renyi_generator.hpp>
+#include <boost/graph/distributed/adjacency_list.hpp>
+#include <boost/graph/graphviz.hpp>
+#include <iostream>
+#include <boost/random.hpp>
+#include <boost/test/minimal.hpp>
+
+#ifdef BOOST_NO_EXCEPTIONS
+void
+boost::throw_exception(std::exception const& ex)
+{
+    std::cout << ex.what() << std::endl;
+    abort();
+}
+#endif
+
+using namespace boost;
+using boost::graph::distributed::mpi_process_group;
+
+void 
+test_distributed_graph_coloring(int n, double p, int s,
+                                int seed, bool emit_dot_file)
+{
+  typedef adjacency_list<listS, 
+                         distributedS<mpi_process_group, vecS>,
+                         undirectedS> Graph;
+
+  typedef property_map<Graph, vertex_index_t>::type vertex_index_map;
+
+  // Build a random number generator
+  minstd_rand gen;
+  gen.seed(seed);
+
+  // Build a random graph
+  Graph g(erdos_renyi_iterator<minstd_rand, Graph>(gen, n, p),
+          erdos_renyi_iterator<minstd_rand, Graph>(),
+          n);
+
+  // Set up color map
+  std::vector<int> colors_vec(num_vertices(g));
+  iterator_property_map<int*, vertex_index_map> color(&colors_vec[0],
+                                                      get(vertex_index, g));
+
+  // Run the graph coloring algorithm
+  graph::boman_et_al_graph_coloring(g, color, s);
+
+  if (process_id(g.process_group()) == 0) {
+    graph::distributed::boman_et_al_graph_coloring_stats.print(std::cout);
+  }
+
+  if ( emit_dot_file ) {
+    if ( process_id(g.process_group()) == 0 ) {
+      for (int i = 0; i < n; ++i)
+        get(color, vertex(i, g));
+      synchronize(color);
+    } else {
+      synchronize(color);
+    }
+    
+    write_graphviz("coloring.dot", g, paint_by_number(color));
+  }
+}
+
+int test_main(int argc, char* argv[])
+{
+  mpi::environment env(argc, argv);
+
+  int n = 1000;
+  double p = 0.01;
+  int s = 100;
+  int seed = 1;
+  bool emit_dot_file = false;
+
+  if (argc > 1) n             = lexical_cast<int>(argv[1]);
+  if (argc > 2) p             = lexical_cast<double>(argv[2]);
+  if (argc > 3) s             = lexical_cast<int>(argv[3]);
+  if (argc > 4) seed          = lexical_cast<int>(argv[4]);
+  if (argc > 5) emit_dot_file = lexical_cast<bool>(argv[5]);
+
+  test_distributed_graph_coloring(n, p, s, seed, emit_dot_file);
+
+  return 0;
+}
diff --git a/src/boost/libs/graph_parallel/test/distributed_mst_test.cpp b/src/boost/libs/graph_parallel/test/distributed_mst_test.cpp
new file mode 100644
index 00000000..614116f6
--- /dev/null
+++ b/src/boost/libs/graph_parallel/test/distributed_mst_test.cpp
@@ -0,0 +1,151 @@
+// Copyright 2004 The Trustees of Indiana University.
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+//  Authors: Douglas Gregor
+//           Andrew Lumsdaine
+
+#include <boost/graph/use_mpi.hpp>
+#include <boost/config.hpp>
+#include <boost/throw_exception.hpp>
+#include <boost/serialization/vector.hpp>
+#include <boost/graph/distributed/dehne_gotz_min_spanning_tree.hpp>
+#include <boost/graph/distributed/mpi_process_group.hpp>
+#include <boost/graph/distributed/vertex_list_adaptor.hpp>
+#include <boost/graph/parallel/distribution.hpp>
+#include <boost/test/minimal.hpp>
+#include <boost/graph/distributed/adjacency_list.hpp>
+#include <iostream>
+#include <cstdlib>
+
+#ifdef BOOST_NO_EXCEPTIONS
+void
+boost::throw_exception(std::exception const& ex)
+{
+    std::cout << ex.what() << std::endl;
+    abort();
+}
+#endif
+
+using namespace boost;
+using boost::graph::distributed::mpi_process_group;
+
+template<typename Graph, typename WeightMap, typename InputIterator>
+int
+total_weight(const Graph& g, WeightMap weight_map, 
+             InputIterator first, InputIterator last)
+{
+  typedef typename graph_traits<Graph>::vertex_descriptor vertex_descriptor;
+
+  int total_weight = 0;
+  while (first != last) {
+    total_weight += get(weight_map, *first);
+    if (process_id(g.process_group()) == 0) {
+      vertex_descriptor u = source(*first, g);
+      vertex_descriptor v = target(*first, g);
+      std::cout << "(" << g.distribution().global(owner(u), local(u))
+                << ", " << g.distribution().global(owner(v), local(v))
+                << ")\n";
+    }
+    ++first;
+  }
+
+  return total_weight;
+}
+
+void 
+test_distributed_dense_boruvka()
+{
+  typedef adjacency_list<listS, 
+                         distributedS<mpi_process_group, vecS>,
+                         undirectedS,
+                         // Vertex properties
+                         no_property,
+                         // Edge properties
+                         property<edge_weight_t, int> > Graph;
+
+  typedef graph_traits<Graph>::edge_descriptor edge_descriptor;
+
+  typedef std::pair<int, int> E;
+
+  const int num_nodes = 5;
+  E edge_array[] = { E(0, 2), E(1, 3), E(1, 4), E(2, 1), E(2, 3),
+    E(3, 4), E(4, 0), E(4, 1)
+  };
+  int weights[] = { 1, 1, 2, 7, 3, 1, 1, 1 };
+  std::size_t num_edges = sizeof(edge_array) / sizeof(E);
+
+  Graph g(edge_array, edge_array + num_edges, weights, num_nodes);
+
+  {
+    if (process_id(g.process_group()) == 0)
+      std::cerr << "--Dense Boruvka--\n";
+    typedef property_map<Graph, edge_weight_t>::type WeightMap;
+    WeightMap weight_map = get(edge_weight, g);
+    
+    std::vector<edge_descriptor> mst_edges;
+    dense_boruvka_minimum_spanning_tree(make_vertex_list_adaptor(g), 
+                                        weight_map, 
+                                        std::back_inserter(mst_edges));
+    int w = total_weight(g, weight_map, mst_edges.begin(), mst_edges.end());
+    BOOST_CHECK(w == 4);
+    BOOST_CHECK(mst_edges.size() == 4);
+  }
+
+  {
+    if (process_id(g.process_group()) == 0)
+      std::cerr << "--Merge local MSTs--\n";
+    typedef property_map<Graph, edge_weight_t>::type WeightMap;
+    WeightMap weight_map = get(edge_weight, g);
+    
+    std::vector<edge_descriptor> mst_edges;
+    merge_local_minimum_spanning_trees(make_vertex_list_adaptor(g), weight_map,
+                                       std::back_inserter(mst_edges));
+    if (process_id(g.process_group()) == 0) {
+      int w = total_weight(g, weight_map, mst_edges.begin(), mst_edges.end());
+      BOOST_CHECK(w == 4);
+      BOOST_CHECK(mst_edges.size() == 4);
+    }
+  }
+
+  {
+    if (process_id(g.process_group()) == 0)
+      std::cerr << "--Boruvka then Merge--\n";
+    typedef property_map<Graph, edge_weight_t>::type WeightMap;
+    WeightMap weight_map = get(edge_weight, g);
+    
+    std::vector<edge_descriptor> mst_edges;
+    boruvka_then_merge(make_vertex_list_adaptor(g), weight_map,
+                        std::back_inserter(mst_edges));
+    if (process_id(g.process_group()) == 0) {
+      int w = total_weight(g, weight_map, mst_edges.begin(), mst_edges.end());
+      BOOST_CHECK(w == 4);
+      BOOST_CHECK(mst_edges.size() == 4);
+    }
+  }
+
+  {
+    if (process_id(g.process_group()) == 0)
+      std::cerr << "--Boruvka mixed Merge--\n";
+    typedef property_map<Graph, edge_weight_t>::type WeightMap;
+    WeightMap weight_map = get(edge_weight, g);
+    
+    std::vector<edge_descriptor> mst_edges;
+    boruvka_mixed_merge(make_vertex_list_adaptor(g), weight_map,
+                        std::back_inserter(mst_edges));
+    if (process_id(g.process_group()) == 0) {
+      int w = total_weight(g, weight_map, mst_edges.begin(), mst_edges.end());
+      BOOST_CHECK(w == 4);
+      BOOST_CHECK(mst_edges.size() == 4);
+    }
+  }
+}
+
+int test_main(int argc, char** argv)
+{
+  boost::mpi::environment env(argc, argv);
+  test_distributed_dense_boruvka();
+  return 0;
+}
diff --git a/src/boost/libs/graph_parallel/test/distributed_page_rank_test.cpp b/src/boost/libs/graph_parallel/test/distributed_page_rank_test.cpp
new file mode 100644
index 00000000..d7b29457
--- /dev/null
+++ b/src/boost/libs/graph_parallel/test/distributed_page_rank_test.cpp
@@ -0,0 +1,110 @@
+// Copyright 2004 The Trustees of Indiana University.
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+//  Authors: Douglas Gregor
+//           Andrew Lumsdaine
+
+#include <boost/graph/use_mpi.hpp>
+#include <boost/throw_exception.hpp>
+#include <boost/serialization/vector.hpp>
+#include <boost/graph/distributed/page_rank.hpp>
+#include <boost/test/minimal.hpp>
+#include <boost/graph/distributed/adjacency_list.hpp>
+#include <boost/graph/distributed/mpi_process_group.hpp>
+#include <boost/test/minimal.hpp>
+#include <vector>
+#include <iostream>
+#include <stdlib.h>
+
+#ifdef BOOST_NO_EXCEPTIONS
+void
+boost::throw_exception(std::exception const& ex)
+{
+    std::cout << ex.what() << std::endl;
+    abort();
+}
+#endif
+
+using namespace boost;
+using boost::graph::distributed::mpi_process_group;
+
+bool close_to(double x, double y)
+{
+  double diff = x - y;
+  if (diff < 0) diff = -diff;
+  double base = (y == 0? x : y);
+  if (base != 0) return diff / base < 0.01;
+  else return true;
+}
+
+// Make convenient labels for the vertices
+enum vertex_id_t { A, B, C, D, N };
+
+void test_distributed_page_rank(int iterations)
+{
+  using namespace boost::graph;
+
+  // create a typedef for the Graph type
+  typedef adjacency_list<vecS, 
+                         distributedS<mpi_process_group, vecS>,
+                         bidirectionalS 
+                         > Graph;
+  typedef graph_traits<Graph>::vertex_descriptor vertex_descriptor;
+
+  // writing out the edges in the graph
+  typedef std::pair<int, int> Edge;
+  Edge edge_array[] =
+    { Edge(A,B), Edge(A,C), Edge(B,C), Edge(C,A), Edge(D,C) };
+  const int num_edges = sizeof(edge_array)/sizeof(edge_array[0]);
+
+  // declare a graph object
+  Graph g(edge_array, edge_array + num_edges, N);
+
+  std::vector<double> ranks(num_vertices(g));
+
+  page_rank(g,
+            make_iterator_property_map(ranks.begin(),
+                                       get(boost::vertex_index, g)),
+            n_iterations(iterations), 0.85, N);
+  
+  double local_sum = 0.0;
+  for(unsigned int i = 0; i < num_vertices(g); ++i) {
+    std::cout << (char)('A' + g.distribution().global(i)) << " = " 
+              << ranks[i] << std::endl;
+    local_sum += ranks[i];
+  }
+  double sum=0.;
+  boost::mpi::reduce(communicator(g.process_group()),
+                     local_sum, sum, std::plus<double>(), 0);
+  if (process_id(g.process_group()) == 0) {
+    std::cout << "Sum = " << sum << "\n\n";
+    BOOST_CHECK(close_to(sum, 4)); // 1 when alpha=0
+  }
+
+  //   double expected_ranks0[N] = {0.400009, 0.199993, 0.399998, 0.0};
+  double expected_ranks[N] = {1.49011, 0.783296, 1.5766, 0.15};
+  for (int i = 0; i < N; ++i) {
+    vertex_descriptor v = vertex(i, g);
+    if (v != Graph::null_vertex()
+        && owner(v) == process_id(g.process_group())) {
+      BOOST_CHECK(close_to(ranks[local(v)], expected_ranks[i]));
+    }
+  }
+}
+
+int test_main(int argc, char* argv[])
+{
+  mpi::environment env(argc, argv);
+
+  int iterations = 50;
+  if (argc > 1) {
+    iterations = atoi(argv[1]);
+  }
+
+  test_distributed_page_rank(iterations);
+
+  return 0;
+}
diff --git a/src/boost/libs/graph_parallel/test/distributed_property_map_test.cpp b/src/boost/libs/graph_parallel/test/distributed_property_map_test.cpp
new file mode 100644
index 00000000..6304fffa
--- /dev/null
+++ b/src/boost/libs/graph_parallel/test/distributed_property_map_test.cpp
@@ -0,0 +1,356 @@
+// Copyright (C) 2004-2008 The Trustees of Indiana University.
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+//  Authors: Douglas Gregor
+//           Andrew Lumsdaine
+#include <boost/graph/use_mpi.hpp>
+#include <boost/config.hpp>
+#include <boost/throw_exception.hpp>
+#include <boost/graph/distributed/mpi_process_group.hpp>
+#include <boost/property_map/property_map.hpp>
+#include <boost/test/minimal.hpp>
+#include <vector>
+#include <string>
+#include <boost/serialization/vector.hpp>
+#include <boost/serialization/string.hpp>
+#include <boost/serialization/utility.hpp>
+#include <boost/lexical_cast.hpp>
+#include <boost/graph/parallel/basic_reduce.hpp>
+
+#ifdef BOOST_NO_EXCEPTIONS
+void
+boost::throw_exception(std::exception const& ex)
+{
+    std::cout << ex.what() << std::endl;
+    abort();
+}
+#endif
+
+using namespace boost;
+using boost::graph::distributed::mpi_process_group;
+
+enum color_t { red, blue };
+
+struct remote_key 
+{
+  remote_key(int p = -1, std::size_t l = 0) : processor(p), local_key(l) {}
+
+  int processor;
+  std::size_t local_key;
+
+  template<typename Archiver>
+  void serialize(Archiver& ar, const unsigned int /*version*/)
+  {
+    ar & processor & local_key;
+  }
+};
+
+namespace boost { namespace mpi {
+    template<> struct is_mpi_datatype<remote_key> : mpl::true_ { };
+} }
+BOOST_IS_BITWISE_SERIALIZABLE(remote_key)
+BOOST_CLASS_IMPLEMENTATION(remote_key,object_serializable)
+BOOST_CLASS_TRACKING(remote_key,track_never)
+
+namespace boost { 
+
+template<>
+struct hash<remote_key>
+{
+  std::size_t operator()(const remote_key& key) const
+  {
+    std::size_t hash = hash_value(key.processor);
+    hash_combine(hash, key.local_key);
+    return hash;
+  }
+};
+}
+
+inline bool operator==(const remote_key& x, const remote_key& y)
+{ return x.processor == y.processor && x.local_key == y.local_key; }
+
+struct remote_key_to_global
+{
+  typedef readable_property_map_tag category;
+  typedef remote_key key_type;
+  typedef std::pair<int, std::size_t> value_type;
+  typedef value_type reference;
+};
+
+inline std::pair<int, std::size_t> 
+get(remote_key_to_global, const remote_key& key)
+{
+  return std::make_pair(key.processor, key.local_key);
+}
+
+template<typename T>
+struct my_reduce : boost::parallel::basic_reduce<T> {
+  BOOST_STATIC_CONSTANT(bool, non_default_resolver = true);
+};
+
+void colored_test()
+{
+  mpi_process_group pg;
+  const int n = 500;
+  
+  color_t my_start_color = process_id(pg) % 2 == 0? ::red : ::blue;
+  int next_processor = (process_id(pg) + 1) % num_processes(pg);
+  color_t next_start_color = next_processor % 2 == 0? ::red : ::blue;
+
+  // Initial color map: even-numbered processes are all red,
+  // odd-numbered processes are all blue.
+  std::vector<color_t> color_vec(n, my_start_color);
+
+  typedef iterator_property_map<std::vector<color_t>::iterator, 
+                                identity_property_map> LocalPropertyMap;
+  LocalPropertyMap local_colors(color_vec.begin(), identity_property_map());
+
+  synchronize(pg);
+
+  // Create the distributed property map
+  typedef boost::parallel::distributed_property_map<mpi_process_group,
+                                                    remote_key_to_global,
+                                                    LocalPropertyMap> ColorMap;
+  ColorMap colors(pg, remote_key_to_global(), local_colors);
+  colors.set_reduce(my_reduce<color_t>());
+
+  if (process_id(pg) == 0) std::cerr << "Checking local colors...";
+  // check local processor colors
+  for (int i = 0; i < n; ++i) {
+    remote_key k(process_id(pg), i);
+    BOOST_CHECK(get(colors, k) == my_start_color);
+  }
+
+  colors.set_consistency_model(boost::parallel::cm_bidirectional);
+  if (process_id(pg) == 0) std::cerr << "OK.\nChecking next processor's default colors...";
+  // check next processor's colors
+  for (int i = 0; i < n; ++i) {
+    remote_key k(next_processor, i);
+    BOOST_CHECK(get(colors, k) == color_t());
+  }
+
+  if (process_id(pg) == 0) std::cerr << "OK.\nSynchronizing...";
+  synchronize(pg);
+
+  if (process_id(pg) == 0) std::cerr << "OK.\nChecking next processor's colors...";
+  // check next processor's colors
+  for (int i = 0; i < n; ++i) {
+    remote_key k(next_processor, i);
+    BOOST_CHECK(get(colors, k) == next_start_color);
+  }
+
+  if (process_id(pg) == 0) std::cerr << "OK.\nSynchronizing...";
+  synchronize(pg);
+
+  if (process_id(pg) == 0) std::cerr << "OK.\nChanging next processor's colors...";
+  // change the next processor's colors
+  color_t next_finish_color = next_processor % 2 == 0? ::blue : ::red;
+  for (int i = 0; i < n; ++i) {
+    remote_key k(next_processor, i);
+    put(colors, k, next_finish_color);
+  }
+
+  if (process_id(pg) == 0) std::cerr << "OK.\nSynchronizing...";
+  synchronize(pg);
+
+  if (process_id(pg) == 0) std::cerr << "OK.\nChecking changed colors...";
+  // check our own colors
+  color_t my_finish_color = process_id(pg) % 2 == 0? ::blue : ::red;
+  for (int i = 0; i < n; ++i) {
+    remote_key k(process_id(pg), i);
+    BOOST_CHECK(get(colors, k) == my_finish_color);
+  }
+
+  // check our neighbor's colors
+  if (process_id(pg) == 0) std::cerr << "OK.\nChecking changed colors on neighbor...";
+  for (int i = 0; i < n; ++i) {
+    remote_key k(next_processor, i);
+    BOOST_CHECK(get(colors, k) == next_finish_color);
+  }
+
+  synchronize(pg);
+
+  if (process_id(pg) == 0) std::cerr << "OK.\n";
+}
+
+void bool_test()
+{
+  mpi_process_group pg;
+  const int n = 500;
+  
+  bool my_start_value = process_id(pg) % 2;
+  int next_processor = (process_id(pg) + 1) % num_processes(pg);
+  bool next_start_value = ((process_id(pg) + 1) % num_processes(pg)) % 2;
+
+  // Initial color map: even-numbered processes are false, 
+  // odd-numbered processes are true
+  std::vector<bool> bool_vec(n, my_start_value);
+
+  typedef iterator_property_map<std::vector<bool>::iterator, 
+                                identity_property_map> LocalPropertyMap;
+  LocalPropertyMap local_values(bool_vec.begin(), identity_property_map());
+
+  synchronize(pg);
+
+  // Create the distributed property map
+  typedef boost::parallel::distributed_property_map<mpi_process_group,
+                                                    remote_key_to_global,
+                                                    LocalPropertyMap> ValueMap;
+  ValueMap values(pg, remote_key_to_global(), local_values);
+  values.set_reduce(my_reduce<bool>());
+
+  if (process_id(pg) == 0) std::cerr << "Checking local values...";
+  // check local processor values
+  for (int i = 0; i < n; ++i) {
+    remote_key k(process_id(pg), i);
+    BOOST_CHECK(get(values, k) == my_start_value);
+  }
+
+  values.set_consistency_model(boost::parallel::cm_bidirectional);
+  if (process_id(pg) == 0) std::cerr << "OK.\nChecking next processor's default values...";
+  // check next processor's values
+  for (int i = 0; i < n; ++i) {
+    remote_key k(next_processor, i);
+    BOOST_CHECK(get(values, k) == false);
+  }
+
+  if (process_id(pg) == 0) std::cerr << "OK.\nSynchronizing...";
+  synchronize(pg);
+
+  if (process_id(pg) == 0) std::cerr << "OK.\nChecking next processor's values...";
+  // check next processor's values
+  for (int i = 0; i < n; ++i) {
+    remote_key k(next_processor, i);
+    BOOST_CHECK(get(values, k) == next_start_value);
+  }
+
+  if (process_id(pg) == 0) std::cerr << "OK.\nSynchronizing...";
+  synchronize(pg);
+
+  if (process_id(pg) == 0) std::cerr << "OK.\nChanging next processor's values...";
+  // change the next processor's values
+  bool next_finish_value = next_processor % 2 == 0;
+  for (int i = 0; i < n; ++i) {
+    remote_key k(next_processor, i);
+    put(values, k, next_finish_value);
+  }
+
+  if (process_id(pg) == 0) std::cerr << "OK.\nSynchronizing...";
+  synchronize(pg);
+
+  if (process_id(pg) == 0) std::cerr << "OK.\nChecking changed values...";
+  // check our own values
+  bool my_finish_value = process_id(pg) % 2 == 0;
+  for (int i = 0; i < n; ++i) {
+    remote_key k(process_id(pg), i);
+    BOOST_CHECK(get(values, k) == my_finish_value);
+  }
+
+  // check our neighbor's values
+  if (process_id(pg) == 0) std::cerr << "OK.\nChecking changed values on neighbor...";
+  for (int i = 0; i < n; ++i) {
+    remote_key k(next_processor, i);
+    BOOST_CHECK(get(values, k) == next_finish_value);
+  }
+
+  synchronize(pg);
+
+  if (process_id(pg) == 0) std::cerr << "OK.\n";
+}
+
+void string_test()
+{
+  mpi_process_group pg;
+  const int n = 500;
+  
+  std::string my_start_string = lexical_cast<std::string>(process_id(pg));
+  int next_processor = (process_id(pg) + 1) % num_processes(pg);
+  std::string next_start_string = lexical_cast<std::string>(next_processor);
+
+  // Initial color map: even-numbered processes are false, 
+  // odd-numbered processes are true
+  std::vector<std::string> string_vec(n, my_start_string);
+
+  typedef iterator_property_map<std::vector<std::string>::iterator, 
+                                identity_property_map> LocalPropertyMap;
+  LocalPropertyMap local_strings(string_vec.begin(), identity_property_map());
+
+  synchronize(pg);
+
+  // Create the distributed property map
+  typedef boost::parallel::distributed_property_map<mpi_process_group,
+                                                    remote_key_to_global,
+                                                    LocalPropertyMap> StringMap;
+  StringMap strings(pg, remote_key_to_global(), local_strings);
+  strings.set_reduce(my_reduce<std::string>());
+
+  if (process_id(pg) == 0) std::cerr << "Checking local strings...";
+  // check local processor strings
+  for (int i = 0; i < n; ++i) {
+    remote_key k(process_id(pg), i);
+    BOOST_CHECK(get(strings, k) == my_start_string);
+  }
+
+  strings.set_consistency_model(boost::parallel::cm_bidirectional);
+  if (process_id(pg) == 0) std::cerr << "OK.\nChecking next processor's default strings...";
+  // check next processor's strings
+  for (int i = 0; i < n; ++i) {
+    remote_key k(next_processor, i);
+    BOOST_CHECK(get(strings, k) == (num_processes(pg) == 1 ? my_start_string : std::string()));
+  }
+
+  if (process_id(pg) == 0) std::cerr << "OK.\nSynchronizing...";
+  synchronize(pg);
+
+  if (process_id(pg) == 0) std::cerr << "OK.\nChecking next processor's strings...";
+  // check next processor's strings
+  for (int i = 0; i < n; ++i) {
+    remote_key k(next_processor, i);
+    BOOST_CHECK(get(strings, k) == next_start_string);
+  }
+
+  if (process_id(pg) == 0) std::cerr << "OK.\nSynchronizing...";
+  synchronize(pg);
+
+  if (process_id(pg) == 0) std::cerr << "OK.\nChanging next processor's strings...";
+  // change the next processor's strings
+  std::string next_finish_string = next_start_string + next_start_string;
+  for (int i = 0; i < n; ++i) {
+    remote_key k(next_processor, i);
+    put(strings, k, next_finish_string);
+  }
+
+  if (process_id(pg) == 0) std::cerr << "OK.\nSynchronizing...";
+  synchronize(pg);
+
+  if (process_id(pg) == 0) std::cerr << "OK.\nChecking changed strings...";
+  // check our own strings
+  std::string my_finish_string = my_start_string + my_start_string;
+  for (int i = 0; i < n; ++i) {
+    remote_key k(process_id(pg), i);
+    BOOST_CHECK(get(strings, k) == my_finish_string);
+  }
+
+  // check our neighbor's strings
+  if (process_id(pg) == 0) std::cerr << "OK.\nChecking changed strings on neighbor...";
+  for (int i = 0; i < n; ++i) {
+    remote_key k(next_processor, i);
+    BOOST_CHECK(get(strings, k) == next_finish_string);
+  }
+
+  synchronize(pg);
+
+  if (process_id(pg) == 0) std::cerr << "OK.\n";
+}
+
+int test_main(int argc, char** argv)
+{
+  boost::mpi::environment env(argc, argv);
+  colored_test();
+  bool_test();
+  string_test();
+  return 0;
+}
diff --git a/src/boost/libs/graph_parallel/test/distributed_queue_test.cpp b/src/boost/libs/graph_parallel/test/distributed_queue_test.cpp
new file mode 100644
index 00000000..5a363038
--- /dev/null
+++ b/src/boost/libs/graph_parallel/test/distributed_queue_test.cpp
@@ -0,0 +1,116 @@
+// Copyright (C) 2004-2006 The Trustees of Indiana University.
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+//  Authors: Douglas Gregor
+//           Andrew Lumsdaine
+#include <boost/graph/use_mpi.hpp>
+#include <boost/config.hpp>
+#include <boost/throw_exception.hpp>
+#include <boost/serialization/vector.hpp>
+#include <boost/graph/distributed/queue.hpp>
+#include <boost/test/minimal.hpp>
+#include <boost/graph/distributed/mpi_process_group.hpp>
+#include <boost/pending/queue.hpp>
+#include <boost/property_map/property_map.hpp>
+#include <utility>
+#include <iostream>
+
+#ifdef BOOST_NO_EXCEPTIONS
+void
+boost::throw_exception(std::exception const& ex)
+{
+    std::cout << ex.what() << std::endl;
+    abort();
+}
+#endif
+
+using boost::graph::distributed::mpi_process_group;
+
+struct global_value 
+{
+  global_value(int p = -1, std::size_t l = 0) : processor(p), value(l) {}
+
+  int processor;
+  std::size_t value;
+
+  template<typename Archiver>
+  void serialize(Archiver& ar, const unsigned int /*version*/)
+  {
+    ar & processor & value;
+  }
+};
+
+namespace boost { namespace mpi {
+    template<> struct is_mpi_datatype<global_value> : mpl::true_ { };
+} } // end namespace boost::mpi
+
+BOOST_IS_BITWISE_SERIALIZABLE(global_value)
+BOOST_CLASS_IMPLEMENTATION(global_value,object_serializable)
+BOOST_CLASS_TRACKING(global_value,track_never)
+
+inline bool operator==(const global_value& x, const global_value& y)
+{ return x.processor == y.processor && x.value == y.value; }
+
+struct global_value_owner_map
+{
+  typedef int value_type;
+  typedef value_type reference;
+  typedef global_value key_type;
+  typedef boost::readable_property_map_tag category;
+};
+
+int get(global_value_owner_map, global_value k)
+{
+  return k.processor;
+}
+
+void test_distributed_queue()
+{
+  mpi_process_group process_group;
+
+  typedef boost::queue<global_value> local_queue_type;
+
+  typedef boost::graph::distributed::distributed_queue<mpi_process_group, 
+                                             global_value_owner_map,
+                                             local_queue_type> dist_queue_type;
+
+  dist_queue_type Q(process_group, global_value_owner_map());
+
+  mpi_process_group::process_id_type id = process_id(process_group),
+    n = num_processes(process_group);
+  global_value v(0, 0);
+
+  if (id == 0) {
+    std::cerr << "Should print level of each processor in a binary tree:\n";
+  }
+  synchronize(process_group);
+
+  if (id == n-1) Q.push(v);
+  while (!Q.empty()) {
+    v = Q.top(); Q.pop();
+    
+    std::cerr << "#" << id << ": level = " << v.value << std::endl;
+
+    int level_begin = 1;
+    for (std::size_t i = 0; i < v.value; ++i) level_begin *= 2;
+    int level_end = level_begin * 2;
+    BOOST_CHECK(level_begin <= (id + 1));
+    BOOST_CHECK((id + 1) <= level_end);
+
+    ++v.value;
+    v.processor = v.processor * 2 + 1;
+    if (v.processor < n) Q.push(v);
+    ++v.processor;
+    if (v.processor < n) Q.push(v);
+  }
+}
+
+int test_main(int argc, char** argv)
+{
+  boost::mpi::environment env(argc, argv);
+  test_distributed_queue();
+  return 0;
+}
diff --git a/src/boost/libs/graph_parallel/test/distributed_rmat_cc.cpp b/src/boost/libs/graph_parallel/test/distributed_rmat_cc.cpp
new file mode 100644
index 00000000..60638175
--- /dev/null
+++ b/src/boost/libs/graph_parallel/test/distributed_rmat_cc.cpp
@@ -0,0 +1,118 @@
+// Copyright (C) 2006 The Trustees of Indiana University.
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+//  Authors: Brian Barrett
+//           Nick Edmonds
+//           Andrew Lumsdaine
+
+#include <boost/graph/use_mpi.hpp>
+#include <boost/config.hpp>
+#include <boost/throw_exception.hpp>
+#include <boost/graph/distributed/adjacency_list.hpp>
+#include <boost/property_map/parallel/distributed_property_map.hpp>
+#include <boost/graph/distributed/mpi_process_group.hpp>
+#include <boost/graph/distributed/concepts.hpp>
+#include <boost/graph/distributed/connected_components_parallel_search.hpp>
+#include <boost/graph/distributed/connected_components.hpp>
+#include <boost/graph/rmat_graph_generator.hpp>
+#include <boost/random/linear_congruential.hpp>
+#include <boost/graph/graphviz.hpp>
+#include <boost/property_map/vector_property_map.hpp>
+
+#include <iostream>
+#include <cstdlib>
+#include <iomanip>
+
+#ifdef BOOST_NO_EXCEPTIONS
+void
+boost::throw_exception(std::exception const& ex)
+{
+  std::cout << ex.what() << std::endl;
+  abort();
+}
+#endif
+
+using namespace boost;
+using boost::graph::distributed::mpi_process_group;
+
+typedef double time_type;
+
+inline time_type get_time()
+{
+  return MPI_Wtime();
+}
+
+std::string print_time(time_type t)
+{
+  std::ostringstream out;
+  out << std::setiosflags(std::ios::fixed) << std::setprecision(2) << t;
+  return out.str();
+}
+
+void
+test_filtered_rmat_cc(int n, int m, double a, double b, double c, double d)
+{
+  mpi_process_group pg;
+  mpi_process_group::process_id_type id = process_id(pg);
+
+  if (id == 0) printf("INFO: Params: n=%d, m=%d, a=%f, b=%f, c=%f, d=%f.\n",
+                      n, m, a, b, c, d);
+
+  parallel::variant_distribution<mpi_process_group> distrib 
+    = parallel::block(pg, n);
+
+  typedef adjacency_list<vecS, 
+        distributedS<mpi_process_group, vecS>,
+        undirectedS> Graph;
+
+  typedef keep_local_edges<parallel::variant_distribution<mpi_process_group>,
+                           mpi_process_group::process_id_type>
+    EdgeFilter; 
+
+  typedef unique_rmat_iterator<rand48, Graph, EdgeFilter>
+    RMATIter;
+
+  if (id == 0) printf("INFO: Generating graph.\n");
+
+  rand48 gen;
+  Graph g(RMATIter(gen, n, m, a, b, c, d, true, EdgeFilter(distrib, id)), 
+          RMATIter(), n, pg, distrib);
+
+  synchronize(g);
+
+  if (id == 0) printf("INFO: Starting connected components.\n");
+
+  std::vector<int> local_components_vec(num_vertices(g));
+  typedef iterator_property_map<std::vector<int>::iterator, property_map<Graph, vertex_index_t>::type> ComponentMap;
+  ComponentMap component(local_components_vec.begin(), get(vertex_index, g));
+
+  time_type start = get_time();
+  int num_components = connected_components(g, component);
+  time_type end = get_time();
+
+  if (process_id(g.process_group()) == 0) {
+      std::cout << "INFO: Test Complete. components found = " << num_components
+                << ", time = " << print_time(end - start) << "s." << std::endl;
+      printf("RESULT: %d %d %d %f %f %f %f %f\n", 
+             num_processes(pg), n, m, a, b, c, d, end - start);
+  }
+
+}
+
+int
+main(int argc, char* argv[])
+{
+  mpi::environment env(argc, argv);
+
+  if (argc < 7) 
+    test_filtered_rmat_cc(40, 200, 0.58, 0.19, 0.19, 0.04);    
+  else 
+    test_filtered_rmat_cc(atoi(argv[1]), atoi(argv[2]),
+                          atof(argv[3]), atof(argv[4]), 
+                          atof(argv[5]), atof(argv[6]));
+
+  return 0;
+}
diff --git a/src/boost/libs/graph_parallel/test/distributed_rmat_cc_ps.cpp b/src/boost/libs/graph_parallel/test/distributed_rmat_cc_ps.cpp
new file mode 100644
index 00000000..020465c5
--- /dev/null
+++ b/src/boost/libs/graph_parallel/test/distributed_rmat_cc_ps.cpp
@@ -0,0 +1,117 @@
+// Copyright (C) 2006 The Trustees of Indiana University.
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+//  Authors: Brian Barrett
+//           Nick Edmonds
+//           Andrew Lumsdaine
+
+#include <boost/graph/use_mpi.hpp>
+#include <boost/config.hpp>
+#include <boost/throw_exception.hpp>
+#include <boost/graph/distributed/adjacency_list.hpp>
+#include <boost/property_map/parallel/distributed_property_map.hpp>
+#include <boost/graph/distributed/mpi_process_group.hpp>
+#include <boost/graph/distributed/concepts.hpp>
+#include <boost/graph/distributed/connected_components_parallel_search.hpp>
+#include <boost/graph/distributed/connected_components.hpp>
+#include <boost/graph/rmat_graph_generator.hpp>
+#include <boost/random/linear_congruential.hpp>
+#include <boost/graph/graphviz.hpp>
+#include <boost/property_map/vector_property_map.hpp>
+
+#include <iostream>
+#include <cstdlib>
+#include <iomanip>
+
+#ifdef BOOST_NO_EXCEPTIONS
+void
+boost::throw_exception(std::exception const& ex)
+{
+  std::cout << ex.what() << std::endl;
+  abort();
+}
+#endif
+
+using namespace boost;
+using boost::graph::distributed::mpi_process_group;
+
+typedef double time_type;
+
+inline time_type get_time()
+{
+  return MPI_Wtime();
+}
+
+std::string print_time(time_type t)
+{
+  std::ostringstream out;
+  out << std::setiosflags(std::ios::fixed) << std::setprecision(2) << t;
+  return out.str();
+}
+
+void
+test_filtered_rmat_cc(int n, int m, double a, double b, double c, double d)
+{
+  mpi_process_group pg;
+  std::size_t id = process_id(pg);
+
+  if (id == 0) printf("INFO: Params: n=%d, m=%d, a=%f, b=%f, c=%f, d=%f.\n",
+                      n, m, a, b, c, d);
+
+  typedef parallel::variant_distribution<mpi_process_group> Distribution;
+  Distribution distrib = parallel::block(pg, n);
+
+  typedef adjacency_list<vecS, 
+        distributedS<mpi_process_group, vecS>,
+        undirectedS> Graph;
+
+  typedef scalable_rmat_iterator<mpi_process_group, Distribution, rand48, Graph>
+    RMATIter;
+
+  if (id == 0) printf("INFO: Generating graph.\n");
+
+  rand48 gen;
+  time_type gen_start = get_time();
+  Graph g(RMATIter(pg, distrib, gen, n, m, a, b, c, d, true),
+          RMATIter(), n, pg, distrib);
+  time_type gen_end = get_time();
+  std::cout << "INFO: Graph Gen time: " << print_time(gen_end - gen_start) << std::endl;
+
+  synchronize(g);
+
+  if (id == 0) printf("INFO: Starting connected components.\n");
+
+  std::vector<int> local_components_vec(num_vertices(g));
+  typedef iterator_property_map<std::vector<int>::iterator, property_map<Graph, vertex_index_t>::type> ComponentMap;
+  ComponentMap component(local_components_vec.begin(), get(vertex_index, g));
+
+  time_type start = get_time();
+  int num_components = connected_components_ps(g, component);
+  time_type end = get_time();
+
+  if (process_id(g.process_group()) == 0) {
+      std::cout << "INFO: Test Complete. components found = " << num_components
+                << ", time = " << print_time(end - start) << "s." << std::endl;
+      printf("RESULT: %d %d %d %f %f %f %f %f\n", 
+             num_processes(pg), n, m, a, b, c, d, end - start);
+  }
+
+}
+
+int
+main(int argc, char* argv[])
+{
+  mpi::environment env(argc, argv);
+
+  if (argc < 7) 
+    test_filtered_rmat_cc(40, 200, 0.58, 0.19, 0.19, 0.04);    
+  else 
+    test_filtered_rmat_cc(atoi(argv[1]), atoi(argv[2]),
+                          atof(argv[3]), atof(argv[4]), 
+                          atof(argv[5]), atof(argv[6]));
+
+  return 0;
+}
diff --git a/src/boost/libs/graph_parallel/test/distributed_rmat_pagerank.cpp b/src/boost/libs/graph_parallel/test/distributed_rmat_pagerank.cpp
new file mode 100644
index 00000000..8700249e
--- /dev/null
+++ b/src/boost/libs/graph_parallel/test/distributed_rmat_pagerank.cpp
@@ -0,0 +1,113 @@
+// Copyright (C) 2006 The Trustees of Indiana University.
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+//  Authors: Brian Barrett
+//           Nick Edmonds
+//           Andrew Lumsdaine
+
+#include <boost/graph/use_mpi.hpp>
+#include <boost/config.hpp>
+#include <boost/throw_exception.hpp>
+#include <boost/graph/distributed/adjacency_list.hpp>
+#include <boost/graph/distributed/compressed_sparse_row_graph.hpp>
+#include <boost/property_map/parallel/distributed_property_map.hpp>
+#include <boost/graph/distributed/mpi_process_group.hpp>
+#include <boost/graph/distributed/concepts.hpp>
+#include <boost/graph/distributed/page_rank.hpp>
+#include <boost/graph/rmat_graph_generator.hpp>
+#include <boost/random/linear_congruential.hpp>
+#include <boost/graph/graphviz.hpp>
+#include <boost/property_map/vector_property_map.hpp>
+
+#include <iostream>
+#include <cstdlib>
+#include <iomanip>
+
+#ifdef BOOST_NO_EXCEPTIONS
+void
+boost::throw_exception(std::exception const& ex)
+{
+  std::cout << ex.what() << std::endl;
+  abort();
+}
+#endif
+
+using namespace boost;
+using boost::graph::distributed::mpi_process_group;
+
+typedef double time_type;
+
+inline time_type get_time()
+{
+  return MPI_Wtime();
+}
+
+std::string print_time(time_type t)
+{
+  std::ostringstream out;
+  out << std::setiosflags(std::ios::fixed) << std::setprecision(2) << t;
+  return out.str();
+}
+
+void
+test_filtered_rmat_pagerank(int n, int m, double a, double b, double c, double d, int iters)
+{
+  mpi_process_group pg;
+  std::size_t id = process_id(pg);
+
+  if (id == 0) printf("INFO: Params: n=%d, m=%d, a=%f, b=%f, c=%f, d=%f, iters=%d.\n",
+                      n, m, a, b, c, d, iters);
+
+  typedef parallel::variant_distribution<mpi_process_group> Distribution;
+  Distribution distrib = parallel::block(pg, n);
+
+  typedef adjacency_list<vecS, 
+        distributedS<mpi_process_group, vecS>,
+        bidirectionalS> Graph;
+
+  typedef scalable_rmat_iterator<mpi_process_group, Distribution, rand48, Graph>
+    RMATIter;
+
+  if (id == 0) printf("INFO: Generating graph.\n");
+
+  rand48 gen;
+  Graph g(RMATIter(pg, distrib, gen, n, m, a, b, c, d, true), 
+          RMATIter(), n, pg, distrib);
+
+  synchronize(g);
+
+  if (id == 0) printf("INFO: Starting PageRank.\n");
+
+  std::vector<double> ranks(num_vertices(g));
+
+  time_type start = get_time();
+  page_rank(g, make_iterator_property_map(ranks.begin(), get(boost::vertex_index, g)),
+            graph::n_iterations(iters), 0.85, n);
+  time_type end = get_time();
+
+  if (process_id(g.process_group()) == 0) {
+      std::cout << "INFO: Test Complete. time = " << 
+          print_time(end - start) << "s." << std::endl;
+      printf("RESULT: %d %d %d %f %f %f %f %f\n", 
+             num_processes(pg), n, m, a, b, c, d, end - start);
+  }
+
+}
+
+int
+main(int argc, char* argv[])
+{
+  mpi::environment env(argc, argv);
+
+  if (argc < 8) 
+    test_filtered_rmat_pagerank(40, 200, 0.58, 0.19, 0.19, 0.04, 10);
+  else
+    test_filtered_rmat_pagerank(atoi(argv[1]), atoi(argv[2]), atof(argv[3]), 
+                                atof(argv[4]), atof(argv[5]), atof(argv[6]),
+                                atoi(argv[7]));
+
+  return 0;
+}
diff --git a/src/boost/libs/graph_parallel/test/distributed_shortest_paths_test.cpp b/src/boost/libs/graph_parallel/test/distributed_shortest_paths_test.cpp
new file mode 100644
index 00000000..fce4c255
--- /dev/null
+++ b/src/boost/libs/graph_parallel/test/distributed_shortest_paths_test.cpp
@@ -0,0 +1,214 @@
+// Copyright (C) 2005, 2006 The Trustees of Indiana University.
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+//  Authors: Douglas Gregor
+//           Andrew Lumsdaine
+#define PBGL_ACCOUNTING
+
+#include <boost/graph/use_mpi.hpp>
+#include <boost/config.hpp>
+#include <boost/throw_exception.hpp>
+#include <boost/graph/distributed/delta_stepping_shortest_paths.hpp>
+#include <boost/graph/distributed/mpi_process_group.hpp>
+#include <boost/lexical_cast.hpp>
+#include <boost/graph/parallel/distribution.hpp>
+#include <boost/graph/erdos_renyi_generator.hpp>
+#include <boost/graph/distributed/adjacency_list.hpp>
+#include <boost/graph/graphviz.hpp>
+#include <boost/random.hpp>
+#include <boost/test/minimal.hpp>
+#include <boost/graph/iteration_macros.hpp>
+
+#include <iostream>
+#include <iomanip>
+
+#ifdef BOOST_NO_EXCEPTIONS
+void
+boost::throw_exception(std::exception const& ex)
+{
+    std::cout << ex.what() << std::endl;
+    abort();
+}
+#endif
+
+/****************************************************************************
+ * Timing                                                                   *
+ ****************************************************************************/
+typedef double time_type;
+
+inline time_type get_time()
+{
+  return MPI_Wtime();
+}
+
+std::string print_time(time_type t)
+{
+  std::ostringstream out;
+  out << std::setiosflags(std::ios::fixed) << std::setprecision(2) << t;
+  return out.str();
+}
+
+/****************************************************************************
+ * Edge weight generator iterator                                           *
+ ****************************************************************************/
+template<typename F, typename RandomGenerator>
+class generator_iterator
+{
+public:
+  typedef std::input_iterator_tag iterator_category;
+  typedef typename F::result_type value_type;
+  typedef const value_type&       reference;
+  typedef const value_type*       pointer;
+  typedef void                    difference_type;
+
+  explicit 
+  generator_iterator(RandomGenerator& gen, const F& f = F()) 
+    : f(f), gen(&gen) 
+  { 
+    value = this->f(gen); 
+  }
+
+  reference operator*() const  { return value; }
+  pointer   operator->() const { return &value; }
+
+  generator_iterator& operator++()
+  {
+    value = f(*gen);
+    return *this;
+  }
+
+  generator_iterator operator++(int)
+  {
+    generator_iterator temp(*this);
+    ++(*this);
+    return temp;
+  }
+
+  bool operator==(const generator_iterator& other) const
+  { return f == other.f; }
+
+  bool operator!=(const generator_iterator& other) const
+  { return !(*this == other); }
+
+private:
+  F f;
+  RandomGenerator* gen;
+  value_type value;
+};
+
+template<typename F, typename RandomGenerator>
+inline generator_iterator<F, RandomGenerator> 
+make_generator_iterator( RandomGenerator& gen, const F& f)
+{ return generator_iterator<F, RandomGenerator>(gen, f); }
+
+/****************************************************************************
+ * Verification                                                             *
+ ****************************************************************************/
+template <typename Graph, typename DistanceMap, typename WeightMap>
+void
+verify_shortest_paths(const Graph& g, DistanceMap distance, 
+                      const WeightMap& weight) {
+
+  distance.set_max_ghost_cells(0);
+
+  BGL_FORALL_VERTICES_T(v, g, Graph) {
+    BGL_FORALL_OUTEDGES_T(v, e, g, Graph) {
+      get(distance, target(e, g));
+    }
+  }
+
+  synchronize(process_group(g));
+
+  BGL_FORALL_VERTICES_T(v, g, Graph) {
+    BGL_FORALL_OUTEDGES_T(v, e, g, Graph) {
+      assert(get(distance, target(e, g)) <= 
+             get(distance, source(e, g)) + get(weight, e));
+    }
+  }
+}
+
+using namespace boost;
+using boost::graph::distributed::mpi_process_group;
+
+typedef int weight_type;
+
+struct WeightedEdge {
+  WeightedEdge(weight_type weight = 0) : weight(weight) { }
+  
+  weight_type weight;
+
+  template<typename Archiver>
+  void serialize(Archiver& ar, const unsigned int /*version*/)
+  {
+    ar & weight;
+  }
+};
+
+struct VertexProperties {
+  VertexProperties(int d = 0)
+    : distance(d) { }
+
+  int distance;
+
+  template<typename Archiver>
+  void serialize(Archiver& ar, const unsigned int /*version*/)
+  {
+    ar & distance;
+  }
+};
+
+void 
+test_distributed_shortest_paths(int n, double p, int c, int seed)
+{
+  typedef adjacency_list<listS, 
+                         distributedS<mpi_process_group, vecS>,
+                         undirectedS,
+                         VertexProperties,
+                         WeightedEdge> Graph;
+
+  typedef graph_traits<Graph>::vertices_size_type vertices_size_type;
+
+  // Build a random number generator
+  minstd_rand gen;
+  gen.seed(seed);
+
+  // Build a random graph
+  Graph g(erdos_renyi_iterator<minstd_rand, Graph>(gen, n, p),
+          erdos_renyi_iterator<minstd_rand, Graph>(),
+          make_generator_iterator(gen, uniform_int<int>(0, c)),
+          n);
+
+  uniform_int<vertices_size_type> rand_vertex(0, n);
+
+  graph::distributed::delta_stepping_shortest_paths(g, 
+                                                    vertex(rand_vertex(gen), g),
+                                                    dummy_property_map(),
+                                                    get(&VertexProperties::distance, g),
+                                                    get(&WeightedEdge::weight, g));
+
+  verify_shortest_paths(g, 
+                        get(&VertexProperties::distance, g), 
+                        get(&WeightedEdge::weight, g));
+}
+
+int test_main(int argc, char* argv[])
+{
+  mpi::environment env(argc, argv);
+
+  int n = 1000;
+  double p = 0.01;
+  int c = 100;
+  int seed = 1;
+
+  if (argc > 1) n             = lexical_cast<int>(argv[1]);
+  if (argc > 2) p             = lexical_cast<double>(argv[2]);
+  if (argc > 3) c             = lexical_cast<int>(argv[3]);
+  if (argc > 4) seed          = lexical_cast<int>(argv[4]);
+
+  test_distributed_shortest_paths(n, p, c, seed);
+
+  return 0;
+}
diff --git a/src/boost/libs/graph_parallel/test/distributed_st_connected_test.cpp b/src/boost/libs/graph_parallel/test/distributed_st_connected_test.cpp
new file mode 100644
index 00000000..a65f404f
--- /dev/null
+++ b/src/boost/libs/graph_parallel/test/distributed_st_connected_test.cpp
@@ -0,0 +1,85 @@
+// Copyright (C) 2004-2006 The Trustees of Indiana University.
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+//  Authors: Nick Edmonds
+//           Andrew Lumsdaine
+
+#include <boost/graph/use_mpi.hpp>
+#include <boost/config.hpp>
+#include <boost/throw_exception.hpp>
+#include <boost/graph/distributed/adjacency_list.hpp>
+#include <boost/graph/random.hpp>
+#include <boost/property_map/parallel/distributed_property_map.hpp>
+#include <boost/graph/distributed/mpi_process_group.hpp>
+#include <boost/graph/distributed/st_connected.hpp>
+#include <boost/graph/parallel/distribution.hpp>
+#include <boost/graph/small_world_generator.hpp>
+#include <iostream>
+#include <cstdlib>
+#include <iomanip>
+#include <boost/random.hpp>
+#include <boost/test/minimal.hpp>
+
+#ifdef BOOST_NO_EXCEPTIONS
+void
+boost::throw_exception(std::exception const& ex)
+{
+    std::cout << ex.what() << std::endl;
+    abort();
+}
+#endif
+
+using namespace boost;
+using boost::graph::distributed::mpi_process_group;
+
+// Set up the vertex names
+enum vertex_id_t { u, v, w, x, y, z, N };
+char vertex_names[] = { 'u', 'v', 'w', 'x', 'y', 'z' };
+
+void 
+test_distributed_st_connected() {
+
+  typedef adjacency_list<listS,
+                         distributedS<mpi_process_group, vecS>,
+                         undirectedS,
+                         // Vertex properties
+                         property<vertex_color_t, default_color_type> >
+    Graph;
+
+  // Specify the edges in the graph
+  {
+    typedef std::pair<int, int> E;
+    E edge_array[] = { E(u, u), E(u, v), E(u, w), E(v, w), E(x, y), 
+                       E(x, z), E(z, y), E(z, z) };
+    Graph g(edge_array, edge_array + sizeof(edge_array) / sizeof(E), N);
+
+    bool connected = st_connected(g, vertex(u, g), vertex(z, g), 
+                                  get(vertex_color, g), get(vertex_owner, g));
+
+    assert(!connected);
+  }
+
+  {
+    typedef std::pair<int, int> E;
+    E edge_array[] = { E(u, v), E(u, w), E(u, x), E(x, v), E(y, x),
+                       E(v, y), E(w, y), E(w, z), E(z, z) };
+    Graph g(edge_array, edge_array + sizeof(edge_array) / sizeof(E), N);
+
+    bool connected = st_connected(g, vertex(u, g), vertex(z, g), 
+                                  get(vertex_color, g), get(vertex_owner, g));
+
+    assert(connected);
+  }
+
+
+}
+
+int test_main(int argc, char* argv[])
+{
+  mpi::environment env(argc, argv);
+  test_distributed_st_connected();
+  return 0;
+}
diff --git a/src/boost/libs/graph_parallel/test/distributed_strong_components_test.cpp b/src/boost/libs/graph_parallel/test/distributed_strong_components_test.cpp
new file mode 100644
index 00000000..cbbd41b5
--- /dev/null
+++ b/src/boost/libs/graph_parallel/test/distributed_strong_components_test.cpp
@@ -0,0 +1,175 @@
+// Copyright (C) 2004-2008 The Trustees of Indiana University.
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+//  Authors: Nick Edmonds
+//           Douglas Gregor
+//           Andrew Lumsdaine
+
+// SCC won't work with CSR currently due to the way the reverse graph 
+// is constructed in the SCC algorithm
+
+#include <boost/graph/use_mpi.hpp>
+
+// #define CSR
+
+#ifdef CSR
+#  include <boost/graph/distributed/compressed_sparse_row_graph.hpp>
+#else
+#  include <boost/graph/distributed/adjacency_list.hpp>
+#endif
+
+#include <boost/config.hpp>
+#include <boost/throw_exception.hpp>
+#include <boost/graph/strong_components.hpp>
+#include <boost/graph/random.hpp>
+#include <boost/property_map/parallel/distributed_property_map.hpp>
+#include <boost/graph/distributed/mpi_process_group.hpp>
+#include <boost/graph/parallel/distribution.hpp>
+#include <boost/graph/erdos_renyi_generator.hpp>
+#include <boost/test/minimal.hpp>
+#include <boost/graph/distributed/graphviz.hpp>
+#include <iostream>
+#include <cstdlib>
+#include <iomanip>
+#include <boost/random.hpp>
+#include <boost/test/minimal.hpp>
+
+#ifdef BOOST_NO_EXCEPTIONS
+void
+boost::throw_exception(std::exception const& ex)
+{
+    std::cout << ex.what() << std::endl;
+    abort();
+}
+#endif
+
+typedef double time_type;
+
+inline time_type get_time()
+{
+  return MPI_Wtime();
+}
+
+std::string print_time(time_type t)
+{
+  std::ostringstream out;
+  out << std::setiosflags(std::ios::fixed) << std::setprecision(2) << t;
+  return out.str();
+}
+
+using namespace boost;
+using boost::graph::distributed::mpi_process_group;
+
+void 
+test_distributed_strong_components(int n, double _p, bool verify, bool emit_dot_file, int seed)
+{
+#ifdef CSR
+  typedef compressed_sparse_row_graph<directedS, no_property, no_property, no_property,
+    distributedS<mpi_process_group> > Graph;
+#else
+  typedef adjacency_list<listS, 
+                         distributedS<mpi_process_group, vecS>,
+                         bidirectionalS > Graph;
+#endif
+
+  minstd_rand gen;
+
+  gen.seed(seed);
+
+  mpi_process_group pg;
+  parallel::variant_distribution<mpi_process_group> distrib 
+    = parallel::block(pg, n);
+
+#ifdef CSR
+  Graph g(sorted_erdos_renyi_iterator<minstd_rand, Graph>(gen, n, _p/2),
+          sorted_erdos_renyi_iterator<minstd_rand, Graph>(),
+          n, pg, distrib);
+#else
+  Graph g(erdos_renyi_iterator<minstd_rand, Graph>(gen, n, _p/2),
+          erdos_renyi_iterator<minstd_rand, Graph>(),
+          n, pg, distrib);
+#endif
+
+  synchronize(g);
+
+  std::vector<int> local_components_vec(num_vertices(g));
+  typedef iterator_property_map<std::vector<int>::iterator, property_map<Graph, vertex_index_t>::type> ComponentMap;
+  ComponentMap component(local_components_vec.begin(), get(vertex_index, g));
+
+  int num_components = 0;
+
+  time_type start = get_time();
+  num_components = strong_components(g, component);
+  time_type end = get_time();
+  if (process_id(g.process_group()) == 0)
+    std::cerr << "Erdos-Reyni graph:\n" << n << " Vertices   " << _p << " Edge probability  " 
+              << num_processes(pg) << " Processors\n"
+              << "Strong Components time = " << print_time(end - start) << " seconds.\n"
+              << num_components << " components identified\n";
+  
+
+  if ( verify )
+    {
+      if ( process_id(g.process_group()) == 0 )
+        {
+          for (int i = 0; i < n; ++i)
+            get(component, vertex(i, g));
+          synchronize(component);
+
+          // Check against the sequential version
+          typedef adjacency_list<listS, vecS, directedS> Graph2;
+          
+          gen.seed(seed);
+          
+          Graph2 g2(erdos_renyi_iterator<minstd_rand, Graph>(gen, n, _p/2),
+                    erdos_renyi_iterator<minstd_rand, Graph>(),
+                    n);
+
+          std::vector<int> component2(n);
+          int seq_num_components = strong_components(g2, make_iterator_property_map(component2.begin(), get(vertex_index, g2)));
+
+          assert(num_components == seq_num_components);
+
+          // Make sure components and component2 match
+          std::map<int, int> c2c;
+          int i;
+          for ( i = 0; i < n; i++ )
+            if ( c2c.find( get(component, vertex(i, g)) ) == c2c.end() )
+              c2c[get(component, vertex(i, g))] = component2[i];
+            else
+              if ( c2c[get(component, vertex(i, g))] != component2[i] )
+                break;
+          
+          if ( i < n )
+            std::cerr << "Unable to verify SCC result...\n";
+          else
+            std::cerr << "Passed verification... " << seq_num_components << " strong components\n"; 
+        }
+      else 
+        {
+          synchronize(component);
+        }
+      if ( emit_dot_file )
+        write_graphviz("scc.dot", g, paint_by_number(component));
+    }
+}
+
+int test_main(int argc, char* argv[])
+{
+  mpi::environment env(argc, argv);
+
+  if (argc == 1)
+    test_distributed_strong_components(10000, 0.0005, true, false, 1);
+  else if ( argc < 5 )
+    std::cerr << "usage: test_distributed_strong_components <int num_vertices> <double p> <bool verify?> <bool emit_dotfile?> [seed]\n";
+  else
+    test_distributed_strong_components
+      (atoi(argv[1]), atof(argv[2]), 
+       argv[3]==std::string("true"), argv[4]==std::string("true"),
+       argc == 5? 1 : atoi(argv[5]));
+
+  return 0;
+}
diff --git a/src/boost/libs/graph_parallel/test/hohberg_biconnected_components_test.cpp b/src/boost/libs/graph_parallel/test/hohberg_biconnected_components_test.cpp
new file mode 100644
index 00000000..f0133300
--- /dev/null
+++ b/src/boost/libs/graph_parallel/test/hohberg_biconnected_components_test.cpp
@@ -0,0 +1,175 @@
+// Copyright (C) 2005-2008 The Trustees of Indiana University.
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+//  Authors: Douglas Gregor
+//           Andrew Lumsdaine
+//
+// Test of Hohberg's distributed biconnected components algorithm.
+#include <boost/graph/use_mpi.hpp>
+#include <boost/config.hpp>
+#include <boost/throw_exception.hpp>
+#include <boost/graph/distributed/hohberg_biconnected_components.hpp>
+#include <boost/graph/distributed/mpi_process_group.hpp>
+#include <boost/graph/distributed/adjacency_list.hpp>
+#include <boost/test/minimal.hpp>
+
+#ifdef BOOST_NO_EXCEPTIONS
+void
+boost::throw_exception(std::exception const& ex)
+{
+    std::cout << ex.what() << std::endl;
+    abort();
+}
+#endif
+
+using boost::graph::distributed::mpi_process_group;
+
+using namespace boost;
+
+template<typename Graph>
+void check_components(const Graph& g, std::size_t num_comps)
+{
+  std::size_t not_mapped = (std::numeric_limits<std::size_t>::max)();
+  std::vector<std::size_t> color_to_name(num_comps, not_mapped);
+  BGL_FORALL_EDGES_T(e, g, Graph) {
+    BOOST_CHECK(get(edge_color, g, e) < num_comps);
+    if (color_to_name[get(edge_color, g, e)] == not_mapped)
+      color_to_name[get(edge_color, g, e)] = get(edge_name, g, e);
+    BOOST_CHECK(color_to_name[get(edge_color,g,e)] == get(edge_name,g,e));
+
+    if (color_to_name[get(edge_color,g,e)] != get(edge_name,g,e)) {
+      for (std::size_t i = 0; i < color_to_name.size(); ++i)
+        std::cerr << color_to_name[i] << ' ';
+        
+      std::cerr << std::endl;
+
+      std::cerr << color_to_name[get(edge_color,g,e)] << " != "
+                << get(edge_name,g,e) << " on edge " 
+                << local(source(e, g)) << " -> " << local(target(e, g)) 
+                << std::endl;
+    }
+  }
+}
+
+template<typename Graph>
+void 
+test_small_hohberg_biconnected_components(Graph& g, int n, int comps_expected,
+                                          bool single_component = true)
+{
+  using boost::graph::distributed::hohberg_biconnected_components;
+
+  bool is_root = (process_id(process_group(g)) == 0);
+
+  if (single_component) {
+    for (int i = 0; i < n; ++i) {
+      if (is_root) std::cerr << "Testing with leader = " << i << std::endl;
+      
+      // Scramble edge_color_map
+      BGL_FORALL_EDGES_T(e, g, Graph)
+        put(edge_color, g, e, 17);
+      
+      typename graph_traits<Graph>::vertex_descriptor leader = vertex(i, g);
+      int num_comps = 
+        hohberg_biconnected_components(g, get(edge_color, g), &leader, 
+                                       &leader + 1);
+      
+      BOOST_CHECK(num_comps == comps_expected);
+      check_components(g, num_comps);
+    }
+  } 
+
+  if (is_root) std::cerr << "Testing simple interface." << std::endl;
+  synchronize(g);
+
+  // Scramble edge_color_map
+  int i = 0;
+  BGL_FORALL_EDGES_T(e, g, Graph) {
+    ++i;
+    put(edge_color, g, e, 17);
+  }
+  std::cerr << process_id(process_group(g)) << " has "
+            << i << " edges.\n";
+
+  int num_comps = hohberg_biconnected_components(g, get(edge_color, g));
+
+  BOOST_CHECK(num_comps == comps_expected);
+  check_components(g, num_comps);
+}
+
+int test_main(int argc, char* argv[])
+{
+  mpi::environment env(argc, argv);
+
+  typedef adjacency_list<listS,
+                         distributedS<mpi_process_group, vecS>,
+                         undirectedS,
+                         // Vertex properties
+                         no_property,
+                         // Edge properties
+                         property<edge_name_t, std::size_t, 
+                         property<edge_color_t, std::size_t> > > Graph;
+
+  typedef std::pair<int, int> E;
+
+  {
+    // Example 1: A single component with 2 bicomponents
+    E edges_init[] = { E(0, 1), E(0, 2), E(1, 3), E(2, 4), E(3, 4), E(4, 5), 
+                       E(4, 6), E(5, 6) };
+    const int m = sizeof(edges_init) / sizeof(E);
+    std::size_t expected_components[m] = { 0, 0, 0, 0, 0, 1, 1, 1 };
+    const int n = 7;
+    Graph g(&edges_init[0], &edges_init[0] + m, &expected_components[0], n);
+
+    int num_comps_expected = 
+      *std::max_element(&expected_components[0], &expected_components[0] + m)
+      + 1;
+
+    test_small_hohberg_biconnected_components(g, n, num_comps_expected);
+  }
+
+  {
+    // Example 2: A single component with 4 bicomponents
+    E edges_init[] = { E(0, 1), E(1, 2), E(2, 0), E(2, 3), E(3, 4), E(4, 5), 
+                       E(5, 2), E(3, 6), E(6, 7), E(7, 8), E(8, 6) };
+    const int m = sizeof(edges_init) / sizeof(E);
+    std::size_t expected_components[m] = { 0, 0, 0, 1, 1, 1, 1, 2, 3, 3, 3 };
+    const int n = 9;    
+    Graph g(&edges_init[0], &edges_init[0] + m, &expected_components[0], n);
+
+    int num_comps_expected = 
+      *std::max_element(&expected_components[0], &expected_components[0] + m)
+      + 1;
+
+    test_small_hohberg_biconnected_components(g, n, num_comps_expected);
+  }
+
+  {
+    // Example 3: Two components, 6 bicomponents
+    // This is just the concatenation of the two previous graphs.
+    E edges_init[] = { /* Example 1 graph */
+                       E(0, 1), E(0, 2), E(1, 3), E(2, 4), E(3, 4), E(4, 5), 
+                       E(4, 6), E(5, 6),
+                       /* Example 2 graph */
+                       E(7, 8), E(8, 9), E(9, 7), E(9, 10), E(10, 11), 
+                       E(11, 12), E(12, 9), E(10, 13), E(13, 14), E(14, 15), 
+                       E(15, 13) };
+    const int m = sizeof(edges_init) / sizeof(E);
+    std::size_t expected_components[m] = 
+      { /* Example 1 */0, 0, 0, 0, 0, 1, 1, 1,
+        /* Example 2 */2, 2, 2, 3, 3, 3, 3, 4, 5, 5, 5 };
+    const int n = 16;    
+    Graph g(&edges_init[0], &edges_init[0] + m, &expected_components[0], n);
+
+    int num_comps_expected = 
+      *std::max_element(&expected_components[0], &expected_components[0] + m)
+      + 1;
+
+    test_small_hohberg_biconnected_components(g, n, num_comps_expected,
+                                              false);
+  }
+
+  return 0;
+}
diff --git a/src/boost/libs/graph_parallel/test/mesh_generator_test.cpp b/src/boost/libs/graph_parallel/test/mesh_generator_test.cpp
new file mode 100644
index 00000000..160d1b50
--- /dev/null
+++ b/src/boost/libs/graph_parallel/test/mesh_generator_test.cpp
@@ -0,0 +1,133 @@
+// Copyright (C) 2004-2008 The Trustees of Indiana University.
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+//  Authors: Douglas Gregor
+//           Andrew Lumsdaine
+
+#include <boost/graph/use_mpi.hpp>
+#include <boost/config.hpp>
+#include <boost/throw_exception.hpp>
+#include <boost/graph/mesh_graph_generator.hpp>
+#include <boost/test/minimal.hpp>
+#include <boost/graph/distributed/adjacency_list.hpp>
+#include <boost/graph/distributed/mpi_process_group.hpp>
+#include <boost/graph/distributed/graphviz.hpp>
+#include <boost/graph/iteration_macros.hpp>
+#include <iostream>
+#include <sstream>
+#include <iomanip>
+#include <string>
+#include <boost/test/minimal.hpp>
+
+#ifdef BOOST_NO_EXCEPTIONS
+void
+boost::throw_exception(std::exception const& ex)
+{
+    std::cout << ex.what() << std::endl;
+    abort();
+}
+#endif
+
+using namespace boost;
+using boost::graph::distributed::mpi_process_group;
+
+/****************************************************************************
+ * Timing
+ ****************************************************************************/
+typedef double time_type;
+
+inline time_type get_time()
+{
+        return MPI_Wtime();
+}
+
+std::string print_time(time_type t)
+{
+  std::ostringstream out;
+  out << std::setiosflags(std::ios::fixed) << std::setprecision(2) << t;
+  return out.str();
+}
+
+/****************************************************************************
+ * Edge weight generator iterator                                           *
+ ****************************************************************************/
+template<typename F>
+class generator_iterator
+{
+public:
+  typedef std::input_iterator_tag iterator_category;
+  typedef typename F::result_type value_type;
+  typedef const value_type&       reference;
+  typedef const value_type*       pointer;
+  typedef void                    difference_type;
+
+  explicit generator_iterator(const F& f = F()) : f(f) { value = this->f(); }
+
+  reference operator*() const  { return value; }
+  pointer   operator->() const { return &value; }
+
+  generator_iterator& operator++()
+  {
+    value = f();
+    return *this;
+  }
+
+  generator_iterator operator++(int)
+  {
+    generator_iterator temp(*this);
+    ++(*this);
+    return temp;
+  }
+
+  bool operator==(const generator_iterator& other) const
+  { return f == other.f; }
+
+  bool operator!=(const generator_iterator& other) const
+  { return !(*this == other); }
+
+private:
+  F f;
+  value_type value;
+};
+
+template<typename F>
+inline generator_iterator<F> make_generator_iterator(const F& f)
+{ return generator_iterator<F>(f); }
+
+int test_main(int argc, char* argv[])
+{
+  mpi::environment env(argc, argv);
+
+  if (argc < 5) {
+    std::cerr << "Usage: mesh_generator_test <x> <y> <toroidal> <emit dot file>\n";
+    exit(-1);
+  }
+
+  int x(atoi(argv[1])), y(atoi(argv[2]));
+  bool toroidal(argv[3] == std::string("true"));
+  bool emit_dot_file(argv[4] == std::string("true"));
+
+  typedef adjacency_list<listS, 
+                         distributedS<mpi_process_group, vecS>,
+                         undirectedS> Graph;
+
+  Graph g(mesh_iterator<Graph>(x, y, toroidal),
+          mesh_iterator<Graph>(),
+          x*y);
+
+  synchronize(g);
+
+  BGL_FORALL_VERTICES(v, g, Graph)
+    if (toroidal) 
+      assert(out_degree(v, g) == 4);
+    else
+      assert(out_degree(v, g) >= 2 && out_degree(v, g) <= 4);  
+
+  if ( emit_dot_file )
+    write_graphviz("mesh.dot", g);
+
+  return 0;
+}
diff --git a/src/boost/libs/graph_parallel/test/named_vertices_hash_test.cpp b/src/boost/libs/graph_parallel/test/named_vertices_hash_test.cpp
new file mode 100644
index 00000000..c859a9a6
--- /dev/null
+++ b/src/boost/libs/graph_parallel/test/named_vertices_hash_test.cpp
@@ -0,0 +1,158 @@
+// Copyright (C) 2007-2008 The Trustees of Indiana University.
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+#include <boost/graph/use_mpi.hpp>
+#include <boost/config.hpp>
+#include <boost/throw_exception.hpp>
+#include <boost/mpl/print.hpp>
+#include <boost/graph/distributed/adjacency_list.hpp>
+#include <boost/graph/distributed/mpi_process_group.hpp>
+#include <boost/graph/iteration_macros.hpp>
+#include <boost/test/minimal.hpp>
+#include <string>
+#include <iostream>
+
+#ifdef BOOST_NO_EXCEPTIONS
+void
+boost::throw_exception(std::exception const& ex)
+{
+    std::cout << ex.what() << std::endl;
+    abort();
+}
+#endif
+
+using namespace boost;
+using boost::graph::distributed::mpi_process_group;
+
+/// City structure to be attached to each vertex
+struct City {
+  City() {}
+
+  City(const std::string& name, int population = -1)
+    : name(name), population(population) { }
+  
+  template<typename Archiver>
+  void serialize(Archiver& ar, const unsigned int /*version*/)
+  {
+    ar & name & population;
+  }
+
+  std::string name;
+  int population;
+};
+
+/// Our distribution function
+template<typename T>
+struct named_vertices_hashed_distribution
+{
+  template<typename ProcessGroup>
+  named_vertices_hashed_distribution(const ProcessGroup& pg, 
+        std::size_t /*num_vertices*/ = 0)
+    : n(num_processes(pg)) { }
+
+  int operator()(const T& value) const 
+  {
+    return hasher(value) % n;
+  }
+
+  std::size_t n;
+  boost::hash<T> hasher;
+};
+
+typedef named_vertices_hashed_distribution<std::string> hasher_type;
+
+namespace boost { namespace graph {
+
+/// Use the City name as a key for indexing cities in a graph
+template<>
+struct internal_vertex_name<City>
+{
+  typedef multi_index::member<City, std::string, &City::name> type;
+};
+
+/// Allow the graph to build cities given only their names (filling in
+/// the defaults for fields).
+template<>
+struct internal_vertex_constructor<City>
+{
+  typedef vertex_from_name<City> type;
+};
+
+//   namespace distributed
+//   {
+//     /// Use the City name as the source for the distribution hasher
+//     ///
+//     /// This is currently needed in addition to the specification of this
+//     /// hasher functor as the 3rd template parameter to the distributedS 
+//     /// template.
+//     template<>
+//     struct internal_vertex_name_distribution<City>
+//     {
+//       typedef named_vertices_hashed_distribution<std::string> type;
+//     };
+//   }
+
+} } // end namespace boost::graph
+
+/// Our road map, where each of the vertices are cities
+typedef boost::adjacency_list<vecS, 
+                       distributedS<mpi_process_group, vecS, hasher_type>, 
+                       bidirectionalS, City> RoadMap;
+typedef graph_traits<RoadMap>::vertex_descriptor Vertex;
+
+int test_main(int argc, char** argv)
+{
+  boost::mpi::environment env(argc, argv);
+
+  mpi_process_group pg;
+  RoadMap map; // (pg, hasher_type(pg));
+
+  int rank = process_id(pg);
+  bool i_am_root = rank == 0;
+
+  /// Create vertices for Bloomington, Indianapolis, Chicago. Everyone will
+  /// try to do this, but only one of each vertex will be added.
+  Vertex bloomington = add_vertex(City("Bloomington", 69291), map);
+  Vertex chicago = add_vertex(City("Chicago", 9500000), map);
+  Vertex indianapolis = add_vertex(City("Indianapolis", 791926), map);
+
+  BGL_FORALL_VERTICES(city, map, RoadMap)
+    std::cout << rank << ": " << map[city].name << ", population " 
+              << map[city].population << std::endl;
+
+  BOOST_CHECK(*find_vertex("Bloomington", map) == bloomington);
+  BOOST_CHECK(*find_vertex("Indianapolis", map) == indianapolis);
+  BOOST_CHECK(*find_vertex("Chicago", map) == chicago);
+
+  if (i_am_root) {
+    add_edge(bloomington, "Indianapolis", map);
+    add_edge("Indianapolis", chicago, map);
+    add_edge("Indianapolis", "Cincinnati", map);
+  }
+
+  synchronize(map);
+
+  {
+    property_map<RoadMap, std::string City::*>::type
+      city_name = get(&City::name, map);
+
+    BGL_FORALL_EDGES(road, map, RoadMap)
+      std::cout << rank << ": " << get(city_name, source(road, map)) << " -> " 
+                << get(city_name, target(road, map)) << std::endl;
+
+    synchronize(map);
+  }
+
+  // Make sure the vertex for "Cincinnati" was created implicitly
+  Vertex cincinnati = *find_vertex("Cincinnati", map);
+  if (get(get(vertex_owner, map), cincinnati) 
+        == process_id(map.process_group()))
+    BOOST_CHECK(map[cincinnati].population == -1);
+
+  synchronize(map);
+
+  return 0;
+}
diff --git a/src/boost/libs/graph_parallel/test/named_vertices_seq.cpp b/src/boost/libs/graph_parallel/test/named_vertices_seq.cpp
new file mode 100644
index 00000000..219214c0
--- /dev/null
+++ b/src/boost/libs/graph_parallel/test/named_vertices_seq.cpp
@@ -0,0 +1,91 @@
+// Copyright (C) 2007-2008 The Trustees of Indiana University.
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+#include <boost/graph/use_mpi.hpp>
+#include <boost/config.hpp>
+#include <boost/throw_exception.hpp>
+#include <boost/test/minimal.hpp>
+#include <boost/graph/adjacency_list.hpp>
+#include <boost/graph/iteration_macros.hpp>
+#include <string>
+#include <iostream>
+
+#ifdef BOOST_NO_EXCEPTIONS
+void
+boost::throw_exception(std::exception const& ex)
+{
+    std::cout << ex.what() << std::endl;
+    abort();
+}
+#endif
+
+using namespace boost;
+
+/// City structure to be attached to each vertex
+struct City {
+  City() {}
+
+  City(const std::string& name, int population = -1)
+    : name(name), population(population) { }
+  
+  std::string name;
+  int population;
+};
+
+namespace boost { namespace graph {
+
+/// Use the City name as a key for indexing cities in a graph
+template<>
+struct internal_vertex_name<City>
+{
+  typedef multi_index::member<City, std::string, &City::name> type;
+};
+
+/// Allow the graph to build cities given only their names (filling in
+/// the defaults for fields).
+template<>
+struct internal_vertex_constructor<City>
+{
+  typedef vertex_from_name<City> type;
+};
+
+} } // end namespace boost::graph
+
+/// Our road map, where each of the vertices are cities
+typedef adjacency_list<vecS, vecS, directedS, City> RoadMap;
+typedef graph_traits<RoadMap>::vertex_descriptor Vertex;
+
+int test_main(int argc, char* argv[])
+{
+  RoadMap map;
+
+  /// Create vertices for Bloomington, Indianapolis, Chicago
+  Vertex bloomington = add_vertex(City("Bloomington", 69291), map);
+  Vertex indianapolis = add_vertex(City("Indianapolis", 791926), map);
+  Vertex chicago = add_vertex(City("Chicago", 9500000), map);
+
+  BOOST_CHECK(add_vertex(City("Bloomington", 69291), map) == bloomington);
+
+  BGL_FORALL_VERTICES(city, map, RoadMap)
+    std::cout << map[city].name << ", population " << map[city].population
+              << std::endl;
+
+  BOOST_CHECK(*find_vertex("Bloomington", map) == bloomington);
+  BOOST_CHECK(*find_vertex("Indianapolis", map) == indianapolis);
+  BOOST_CHECK(*find_vertex("Chicago", map) == chicago);
+
+  add_edge(bloomington, "Indianapolis", map);
+  add_edge("Indianapolis", chicago, map);
+  add_edge("Indianapolis", "Cincinnati", map);
+
+  BGL_FORALL_EDGES(road, map, RoadMap)
+    std::cout << map[source(road, map)].name << " -> " 
+              << map[target(road, map)].name << std::endl;
+
+  BOOST_CHECK(map[*find_vertex("Cincinnati", map)].population == -1);
+
+  return 0;
+}
diff --git a/src/boost/libs/graph_parallel/test/named_vertices_test.cpp b/src/boost/libs/graph_parallel/test/named_vertices_test.cpp
new file mode 100644
index 00000000..a5d7b273
--- /dev/null
+++ b/src/boost/libs/graph_parallel/test/named_vertices_test.cpp
@@ -0,0 +1,121 @@
+// Copyright (C) 2007 The Trustees of Indiana University.
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+#include <boost/graph/use_mpi.hpp>
+#include <boost/config.hpp>
+#include <boost/throw_exception.hpp>
+#include <boost/graph/distributed/adjacency_list.hpp>
+#include <boost/graph/distributed/mpi_process_group.hpp>
+#include <boost/graph/iteration_macros.hpp>
+#include <boost/test/minimal.hpp>
+#include <string>
+#include <iostream>
+
+#ifdef BOOST_NO_EXCEPTIONS
+void
+boost::throw_exception(std::exception const& ex)
+{
+    std::cout << ex.what() << std::endl;
+    abort();
+}
+#endif
+
+using namespace boost;
+using boost::graph::distributed::mpi_process_group;
+
+/// City structure to be attached to each vertex
+struct City {
+  City() {}
+
+  City(const std::string& name, int population = -1)
+    : name(name), population(population) { }
+  
+  template<typename Archiver>
+  void serialize(Archiver& ar, const unsigned int /*version*/)
+  {
+    ar & name & population;
+  }
+
+  std::string name;
+  int population;
+};
+
+namespace boost { namespace graph {
+
+/// Use the City name as a key for indexing cities in a graph
+template<>
+struct internal_vertex_name<City>
+{
+  typedef multi_index::member<City, std::string, &City::name> type;
+};
+
+/// Allow the graph to build cities given only their names (filling in
+/// the defaults for fields).
+template<>
+struct internal_vertex_constructor<City>
+{
+  typedef vertex_from_name<City> type;
+};
+
+} } // end namespace boost::graph
+
+/// Our road map, where each of the vertices are cities
+typedef boost::adjacency_list<vecS, distributedS<mpi_process_group, vecS>, 
+                       bidirectionalS, City> RoadMap;
+typedef graph_traits<RoadMap>::vertex_descriptor Vertex;
+
+int test_main(int argc, char** argv)
+{
+  boost::mpi::environment env(argc, argv);
+
+  RoadMap map;
+
+  int rank = process_id(mpi_process_group());
+  bool i_am_root = rank == 0;
+
+  /// Create vertices for Bloomington, Indianapolis, Chicago. Everyone will
+  /// try to do this, but only one of each vertex will be added.
+  Vertex bloomington = add_vertex(City("Bloomington", 69291), map);
+  Vertex chicago = add_vertex(City("Chicago", 9500000), map);
+  Vertex indianapolis = add_vertex(City("Indianapolis", 791926), map);
+
+  BGL_FORALL_VERTICES(city, map, RoadMap)
+    std::cout << rank << ": " << map[city].name << ", population " 
+              << map[city].population << std::endl;
+
+  BOOST_CHECK(*find_vertex("Bloomington", map) == bloomington);
+  BOOST_CHECK(*find_vertex("Indianapolis", map) == indianapolis);
+  BOOST_CHECK(*find_vertex("Chicago", map) == chicago);
+
+  if (i_am_root) {
+    add_edge(bloomington, "Indianapolis", map);
+    add_edge("Indianapolis", chicago, map);
+    add_edge("Indianapolis", "Cincinnati", map);
+  }
+
+  synchronize(map);
+
+  {
+    property_map<RoadMap, std::string City::*>::type
+      city_name = get(&City::name, map);
+
+    BGL_FORALL_EDGES(road, map, RoadMap)
+      std::cout << rank << ": " << get(city_name, source(road, map)) << " -> " 
+                << get(city_name, target(road, map)) << std::endl;
+
+    synchronize(map);
+  }
+
+  // Make sure the vertex for "Cincinnati" was created implicitly
+  Vertex cincinnati = *find_vertex("Cincinnati", map);
+  if (get(vertex_owner, map, cincinnati) 
+        == process_id(map.process_group()))
+    BOOST_CHECK(map[cincinnati].population == -1);
+
+  synchronize(map);
+
+  return 0;
+}
diff --git a/src/boost/libs/graph_parallel/test/process_group_serialization.cpp b/src/boost/libs/graph_parallel/test/process_group_serialization.cpp
new file mode 100644
index 00000000..80700751
--- /dev/null
+++ b/src/boost/libs/graph_parallel/test/process_group_serialization.cpp
@@ -0,0 +1,58 @@
+// Copyright (C) 2006 The Trustees of Indiana University.
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+//  Authors: Douglas Gregor
+//           Andrew Lumsdaine
+
+// FIXME: Including because of a missing header in the serialization library.
+// Patch sent to list...
+#include <cassert>
+
+#include <boost/graph/use_mpi.hpp>
+#include <boost/config.hpp>
+#include <boost/throw_exception.hpp>
+#include <boost/serialization/list.hpp>
+#include <boost/graph/distributed/mpi_process_group.hpp>
+#include <boost/test/minimal.hpp>
+
+#ifdef BOOST_NO_EXCEPTIONS
+void
+boost::throw_exception(std::exception const& ex)
+{
+    std::cout << ex.what() << std::endl;
+    abort();
+}
+#endif
+
+using boost::graph::distributed::mpi_process_group;
+
+int test_main(int argc, char** argv)
+{
+  boost::mpi::environment env(argc, argv);
+
+  mpi_process_group pg;
+
+  int seventeen = 17;
+  std::list<int> seventeens(17, 17);
+
+  if (process_id(pg) == 0) {
+    send(pg, 1, 0, seventeen);
+    send(pg, 1, 1, seventeens);
+  }
+  synchronize(pg);
+
+  if (process_id(pg) == 1) {
+    int value;
+    receive(pg, 0, 0, value);
+    BOOST_CHECK(seventeen == value);
+
+    std::list<int> values;
+    receive(pg, 0, 1, values);
+    BOOST_CHECK(seventeens == values);
+  }
+
+  return 0;
+}
diff --git a/src/boost/libs/graph_parallel/test/ssca.cpp b/src/boost/libs/graph_parallel/test/ssca.cpp
new file mode 100644
index 00000000..0930e4d0
--- /dev/null
+++ b/src/boost/libs/graph_parallel/test/ssca.cpp
@@ -0,0 +1,1263 @@
+// Copyright 2004 The Trustees of Indiana University.
+
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+//  Authors: Nick Edmonds
+//           Andrew Lumsdaine
+
+#include <boost/graph/use_mpi.hpp>
+
+#define CSR
+
+#ifdef CSR
+#  include <boost/graph/distributed/compressed_sparse_row_graph.hpp>
+#else
+#  include <boost/graph/distributed/adjacency_list.hpp>
+#endif
+
+#include <boost/test/minimal.hpp>
+#include <boost/graph/distributed/mpi_process_group.hpp>
+#include <boost/graph/distributed/queue.hpp>
+
+#include <boost/graph/parallel/distribution.hpp>
+#include <boost/lexical_cast.hpp>
+#include <boost/bind.hpp>
+#include <sys/time.h>
+#include <time.h>
+
+#include <boost/random.hpp>
+#include <boost/property_map/parallel/distributed_property_map.hpp>
+
+#include <boost/random/linear_congruential.hpp>
+
+#include <boost/graph/distributed/graphviz.hpp>
+#include <boost/graph/graph_traits.hpp>
+#include <boost/graph/iteration_macros.hpp>
+
+#include <boost/graph/parallel/algorithm.hpp>
+#include <boost/graph/breadth_first_search.hpp>
+#include <boost/pending/queue.hpp>
+
+#include <boost/graph/rmat_graph_generator.hpp>
+#include <boost/graph/distributed/betweenness_centrality.hpp>
+#include <boost/graph/distributed/filtered_graph.hpp>
+#include <boost/graph/parallel/container_traits.hpp>
+
+#include <boost/graph/properties.hpp>
+
+#include <algorithm>
+#include <vector>
+#include <string>
+#include <iostream>
+#include <iomanip>
+#include <fstream>
+#include <string>
+#include <sstream>
+#include <stdint.h>
+
+using namespace boost;
+
+// #define DEBUG
+
+typedef rand48 RandomGenerator;
+
+/****************************************************************************
+ * Timing
+ ****************************************************************************/
+#ifndef PBGL_ACCOUNTING 
+
+typedef double time_type;
+
+inline time_type get_time()
+{
+  timeval tp;
+  gettimeofday(&tp, 0);
+  return tp.tv_sec + tp.tv_usec / 1000000.0;
+}
+
+std::string print_time(time_type t)
+{
+  std::ostringstream out;
+  out << std::setiosflags(std::ios::fixed) << std::setprecision(2) << t;
+  return out.str();
+}
+
+#endif // PBGL_ACCOUNTING
+
+/****************************************************************************
+ * Edge weight generator iterator                                           *
+ ****************************************************************************/
+template<typename F, typename RandomGenerator>
+class generator_iterator
+{
+public:
+  typedef std::input_iterator_tag iterator_category;
+  typedef typename F::result_type value_type;
+  typedef const value_type&       reference;
+  typedef const value_type*       pointer;
+  typedef void                    difference_type;
+
+  explicit 
+  generator_iterator(RandomGenerator& gen, const F& f = F()) 
+    : f(f), gen(&gen) 
+  { 
+    value = this->f(gen); 
+  }
+
+  reference operator*() const  { return value; }
+  pointer   operator->() const { return &value; }
+
+  generator_iterator& operator++()
+  {
+    value = f(*gen);
+    return *this;
+  }
+
+  generator_iterator operator++(int)
+  {
+    generator_iterator temp(*this);
+    ++(*this);
+    return temp;
+  }
+
+  bool operator==(const generator_iterator& other) const
+  { return f == other.f; }
+
+  bool operator!=(const generator_iterator& other) const
+  { return !(*this == other); }
+
+private:
+  F f;
+  RandomGenerator* gen;
+  value_type value;
+};
+
+template<typename F, typename RandomGenerator>
+inline generator_iterator<F, RandomGenerator> 
+make_generator_iterator( RandomGenerator& gen, const F& f)
+{ return generator_iterator<F, RandomGenerator>(gen, f); }
+
+template<typename Graph, typename DistanceMap, typename WeightMap, typename ColorMap>
+struct ssca_visitor : bfs_visitor<> 
+{
+  typedef typename property_traits<WeightMap>::value_type Weight;
+  typedef typename property_traits<ColorMap>::value_type ColorValue;
+  typedef color_traits<ColorValue> Color;
+
+  ssca_visitor(DistanceMap& distance, const WeightMap& weight, ColorMap& color,
+               Weight max_)
+    : distance(distance), weight(weight), color(color), max_(max_) {}
+
+  template<typename Edge>
+  void tree_edge(Edge e, const Graph& g)
+  {
+    int new_distance = get(weight, e) == (std::numeric_limits<Weight>::max)() ? 
+      (std::numeric_limits<Weight>::max)() : get(distance, source(e, g)) + get(weight, e);
+
+    put(distance, target(e, g), new_distance);
+    if (new_distance > max_)
+      put(color, target(e, g), Color::black());
+  }
+
+ private:
+  DistanceMap& distance;
+  const WeightMap& weight;
+  ColorMap& color;
+  Weight max_;
+};
+
+// Generate source vertices for approximate BC
+template <typename Graph, typename Buffer>
+void
+generate_sources(const Graph& g, Buffer sources, 
+                 typename graph_traits<Graph>::vertices_size_type num_sources)
+{
+  typedef typename graph_traits<Graph>::vertex_descriptor vertex_descriptor;
+  typedef typename graph_traits<Graph>::vertices_size_type vertices_size_type;
+  typedef typename graph_traits<Graph>::vertex_iterator vertex_iterator;
+
+  typedef typename boost::graph::parallel::process_group_type<Graph>::type 
+    process_group_type;
+  process_group_type pg = g.process_group();
+
+  typename process_group_type::process_id_type id = process_id(pg);
+  typename process_group_type::process_size_type p = num_processes(pg);    
+  
+  // Don't feel like adding a special case for num_sources < p
+  assert(num_sources >= p);
+  
+  minstd_rand gen;
+  uniform_int<vertices_size_type> rand_vertex(0, num_vertices(g) - 1);
+  std::vector<vertex_descriptor> all_sources, local_sources;
+  vertices_size_type local_vertices = vertices_size_type(floor((double)num_sources / p));
+  local_vertices += (id < (num_sources - (p * local_vertices)) ? 1 : 0); 
+  
+  while (local_vertices > 0) {
+    vertex_iterator iter = vertices(g).first;
+    std::advance(iter, rand_vertex(gen));
+    
+    if (out_degree(*iter, g) != 0
+        && std::find(local_sources.begin(), local_sources.end(), *iter) == local_sources.end()) {
+      local_sources.push_back(*iter);
+      --local_vertices;
+    }
+  }
+  all_gather(pg, local_sources.begin(), local_sources.end(), all_sources);
+  std::sort(all_sources.begin(), all_sources.end());
+  for (typename std::vector<vertex_descriptor>::iterator iter = all_sources.begin(); 
+       iter != all_sources.end(); ++iter)
+    sources.push(*iter);
+}
+
+// Kernel 2 - Classify large sets
+template <typename Graph, typename WeightMap>
+void classify_sets(const Graph& g, const WeightMap& weight_map, 
+                   std::vector<std::pair<typename graph_traits<Graph>::vertex_descriptor,
+                                         typename graph_traits<Graph>::vertex_descriptor> > & global_S)
+{
+  typedef typename boost::graph::parallel::process_group_type<Graph>::type 
+    process_group_type;
+  process_group_type pg = g.process_group();
+
+  typedef typename graph_traits<Graph>::vertex_descriptor vertex_descriptor;
+  std::vector<std::pair<vertex_descriptor, vertex_descriptor> > S;
+
+  time_type start = get_time();
+
+#ifdef CSR
+  typedef typename property_map<Graph, vertex_owner_t>::const_type OwnerMap;
+  typedef typename property_map<Graph, vertex_local_t>::const_type LocalMap;
+  OwnerMap owner = get(vertex_owner, g);
+  LocalMap local = get(vertex_local, g);
+#endif
+
+  int max_ = 0;
+  BGL_FORALL_EDGES_T(e, g, Graph) {
+#ifdef CSR
+    if (get(owner, source(e, g)) == process_id(pg)) {
+#endif
+    int w = get(weight_map, e);
+    if (w > max_) {
+      max_ = w;
+      S.clear();
+    }
+
+    if (w >= max_)
+      S.push_back(std::make_pair(source(e, g), target(e, g)));
+#ifdef CSR
+    }
+#endif
+  }
+
+  int global_max = all_reduce(pg, max_, boost::parallel::maximum<int>());
+  if (max_ < global_max)
+    S.clear();
+
+  global_S.clear();
+
+  all_gather(pg, S.begin(), S.end(), global_S);
+  
+  // This is probably unnecessary as long as the sets of edges owned by procs is disjoint
+  std::sort(global_S.begin(), global_S.end());
+  std::unique(global_S.begin(), global_S.end());
+
+  synchronize(pg);
+
+  time_type end = get_time();
+
+  if (process_id(pg) == 0) {
+    std::cerr << "    Distributed Graph: " << print_time(end - start) << std::endl
+              << "      Max int weight = " << global_max << std::endl;
+  }
+}
+
+template <typename ProcessGroup, typename Graph, typename WeightMap, 
+          typename EdgeVector>
+void seq_classify_sets(const ProcessGroup& pg, const Graph& g, 
+                       const WeightMap& weight_map, EdgeVector& S)
+{
+  typedef typename graph_traits<Graph>::edge_descriptor edge_descriptor;
+  typedef typename property_traits<WeightMap>::value_type edge_weight_type;
+
+  time_type start = get_time();
+
+  edge_weight_type max_ = 0;
+  BGL_FORALL_EDGES_T(e, g, Graph) {
+    edge_weight_type w = get(weight_map, e);
+    if (w > max_) {
+      max_ = w;
+      S.clear();
+    }
+
+    if (w >= max_)
+      S.push_back(e);
+  }
+
+  synchronize(pg);
+
+  time_type end = get_time();
+
+  if (process_id(pg) == 0) 
+    std::cerr << "    Non-Distributed Graph: " << print_time(end - start) << std::endl
+              << "      Max int weight = " << max_ << std::endl;
+}
+
+// Kernel 3 - Graph Extraction
+template <typename Graph, typename OwnerMap, typename LocalMap, 
+          typename WeightMap, typename DistanceMap, typename ColorMap,
+          typename EdgeVector>
+void subgraph_extraction(Graph& g, const OwnerMap& owner, const LocalMap& local,
+                         const WeightMap& weight_map, DistanceMap distances,
+                         ColorMap color_map, const EdgeVector& S,
+                         int subGraphEdgeLength)
+{
+  // Nick: I think turning the vertex black when the maximum distance is
+  //       exceeded will prevent BFS from exploring beyond the subgraph.
+  //       Unfortunately we can't run subgraph extraction in parallel
+  //       because the subgraphs may overlap
+
+  typedef typename property_traits<ColorMap>::value_type ColorValue;
+  typedef color_traits<ColorValue> Color;
+
+  typedef typename graph_traits<Graph>::edge_descriptor edge_descriptor;
+  typedef typename graph_traits<Graph>::vertex_descriptor vertex_descriptor;
+
+  typedef typename boost::graph::parallel::process_group_type<Graph>::type 
+    process_group_type;
+
+  typedef boost::graph::distributed::distributed_queue<process_group_type, 
+    OwnerMap, queue<vertex_descriptor> > queue_t;
+
+  process_group_type pg = g.process_group();
+  typename process_group_type::process_id_type id = process_id(pg);
+
+  queue_t Q(pg, owner);
+
+  EdgeVector sources(S.begin(), S.end());
+
+#ifdef DEBUG
+  std::vector<std::vector<vertex_descriptor> > subgraphs;
+#endif
+
+  synchronize(pg);
+
+  typedef typename std::vector<std::pair<vertex_descriptor, vertex_descriptor> >::iterator
+    source_iterator;
+
+  time_type start = get_time();
+  for (source_iterator iter = sources.begin(); iter != sources.end(); ++iter) {
+    // Reinitialize distance and color maps every BFS
+    BGL_FORALL_VERTICES_T(v, g, Graph) { 
+      if (get(owner, v) == id) {
+        local_put(color_map, v, Color::white()); 
+        local_put(distances, v, (std::numeric_limits<int>::max)());
+      }
+    }
+
+    vertex_descriptor u = iter->first, v = iter->second;
+
+    local_put(distances, u, 0);
+    local_put(distances, v, 0);
+
+    while (!Q.empty()) Q.pop();
+    if (get(owner, u) == id)
+      Q.push(u);
+
+    local_put(color_map, u, Color::gray()); 
+
+    breadth_first_search(g, v, Q,
+                         ssca_visitor<Graph, DistanceMap, WeightMap, ColorMap>
+                           (distances, weight_map, color_map, subGraphEdgeLength),
+                         color_map);
+                           
+    // At this point all vertices with distance > 0 in addition to the 
+    // starting vertices compose the subgraph.
+#ifdef DEBUG
+    subgraphs.push_back(std::vector<vertex_descriptor>());    
+    std::vector<vertex_descriptor>& subgraph = subgraphs.back();
+
+    BGL_FORALL_VERTICES_T(v, g, Graph) {
+      if (get(distances, v) < (std::numeric_limits<int>::max)())
+        subgraph.push_back(v);
+    }
+#endif    
+  }
+
+  synchronize(pg);
+
+  time_type end = get_time();
+
+#ifdef DEBUG
+  for (unsigned int i = 0; i < subgraphs.size(); i++) {
+    all_gather(pg, subgraphs[i].begin(), subgraphs[i].end(), subgraphs[i]);
+    std::sort(subgraphs[i].begin(), subgraphs[i].end()); 
+    subgraphs[i].erase(std::unique(subgraphs[i].begin(), subgraphs[i].end()),
+                       subgraphs[i].end()); 
+  }
+
+  if (process_id(pg) == 0)
+    for (int i = 0; abs(i) < subgraphs.size(); i++) {
+      std::cerr << "Subgraph " << i << " :\n";
+      for (int j = 0; abs(j) < subgraphs[i].size(); j++)
+        std::cerr << "  " << get(local, subgraphs[i][j]) << "@" 
+                  << get(owner, subgraphs[i][j]) << std::endl;
+    }
+#endif
+
+  if (process_id(pg) == 0)
+    std::cerr << "    Distributed Graph: " << print_time(end - start) << std::endl;
+}
+
+template <typename ProcessGroup, typename Graph, typename WeightMap, 
+          typename DistanceMap, typename ColorMap, typename EdgeVector>
+void seq_subgraph_extraction(const ProcessGroup& pg, const Graph& g, 
+                             const WeightMap& weight_map, DistanceMap distances,
+                             ColorMap color_map, const EdgeVector& S,
+                             int subGraphEdgeLength)
+{
+  // Nick: I think turning the vertex black when the maximum distance is
+  //       exceeded will prevent BFS from exploring beyond the subgraph.
+
+  using boost::graph::distributed::mpi_process_group;
+
+  typedef typename property_traits<ColorMap>::value_type ColorValue;
+  typedef color_traits<ColorValue> Color;
+
+  typedef typename graph_traits<Graph>::edge_descriptor edge_descriptor;
+  typedef typename graph_traits<Graph>::vertex_descriptor vertex_descriptor;
+
+  boost::queue<vertex_descriptor> Q;
+
+  std::vector<edge_descriptor> sources(S.begin(), S.end());
+
+#ifdef DEBUG
+  std::vector<std::vector<vertex_descriptor> > subgraphs;
+#endif
+
+  synchronize(pg);
+    
+  typedef ProcessGroup process_group_type;
+  
+  typename process_group_type::process_id_type id = process_id(pg);
+  typename process_group_type::process_size_type p = num_processes(pg);    
+
+  time_type start = get_time();
+
+  for (int i = id; i < sources.size(); i += p) {
+
+    // Reinitialize distance and color maps every BFS
+    BGL_FORALL_VERTICES_T(v, g, Graph) { 
+      put(color_map, v, Color::white()); 
+      put(distances, v, (std::numeric_limits<int>::max)());
+    }
+
+    vertex_descriptor u = source(sources[i], g),
+      v = target(sources[i], g);
+
+    put(distances, u, 0);
+    put(distances, v, 0); 
+
+    while (!Q.empty()) Q.pop();
+    Q.push(u);
+
+    put(color_map, u, Color::gray()); 
+
+    breadth_first_search(g, v, Q,
+                         ssca_visitor<Graph, DistanceMap, WeightMap, ColorMap>
+                           (distances, weight_map, color_map, subGraphEdgeLength),
+                         color_map);
+
+#ifdef DEBUG
+    subgraphs.push_back(std::vector<vertex_descriptor>());    
+    std::vector<vertex_descriptor>& subgraph = subgraphs.back();
+
+    BGL_FORALL_VERTICES_T(v, g, Graph) {
+      if (get(distances, v) < (std::numeric_limits<int>::max)())
+        subgraph.push_back(v);
+    }
+#endif    
+  }
+
+  synchronize(pg);
+
+  time_type end = get_time();
+
+#ifdef DEBUG
+  std::vector<vertex_descriptor> ser_subgraphs;
+
+  for (int i = 0; i < subgraphs.size(); ++i) {
+    for (int j = 0; j < subgraphs[i].size(); ++j)
+      ser_subgraphs.push_back(subgraphs[i][j]);
+    ser_subgraphs.push_back(graph_traits<Graph>::null_vertex());
+  }
+
+  all_gather(pg, ser_subgraphs.begin(), ser_subgraphs.end(), ser_subgraphs);
+
+  int i = 0;
+  typename std::vector<vertex_descriptor>::iterator iter(ser_subgraphs.begin());
+
+  while (iter != ser_subgraphs.end()) {
+    std::cerr << "Subgraph " << i << " :\n";
+    while (*iter != graph_traits<Graph>::null_vertex()) {
+      std::cerr << "  " << *iter << std::endl;
+      ++iter;
+    }
+    ++i;
+    ++iter;
+  }
+#endif
+
+  if (process_id(pg) == 0)
+    std::cerr << "    Non-Distributed Graph: " << print_time(end - start) << std::endl;
+}
+
+template <typename ProcessGroup, typename Graph, typename CentralityMap>
+void
+extract_max_bc_vertices(const ProcessGroup& pg, const Graph& g, const CentralityMap& centrality,
+                        std::vector<typename graph_traits<Graph>::vertex_descriptor>& max_bc_vec)
+{
+  using boost::graph::parallel::process_group;
+  using boost::parallel::all_gather;
+  using boost::parallel::all_reduce;
+
+  // Find set of vertices with highest BC score
+  typedef typename property_traits<CentralityMap>::value_type centrality_type;
+  std::vector<centrality_type> max_bc_vertices;
+  centrality_type max_ = 0;
+
+  max_bc_vec.clear();
+
+  BGL_FORALL_VERTICES_T(v, g, Graph) {
+    if (get(centrality, v) == max_)
+      max_bc_vec.push_back(v);
+    else if (get(centrality, v) > max_) {
+      max_ = get(centrality, v);
+      max_bc_vec.clear();
+      max_bc_vec.push_back(v);      
+    }
+  }
+
+  centrality_type global_max = all_reduce(pg, max_, boost::parallel::minimum<int>());
+
+  if (global_max > max_)
+    max_bc_vec.clear();
+
+  all_gather(pg, max_bc_vec.begin(), max_bc_vec.end(), max_bc_vec);
+}
+
+
+// Function object to filter edges divisible by 8
+// EdgeWeightMap::value_type must be integral!
+template <typename EdgeWeightMap>
+struct edge_weight_not_divisible_by_eight {
+  typedef typename property_traits<EdgeWeightMap>::value_type weight_type;
+
+  edge_weight_not_divisible_by_eight() { }
+  edge_weight_not_divisible_by_eight(EdgeWeightMap weight) : m_weight(weight) { }
+  template <typename Edge>
+  bool operator()(const Edge& e) const {
+    return (get(m_weight, e) & ((std::numeric_limits<weight_type>::max)() - 7)) != get(m_weight, e);
+  }
+
+  EdgeWeightMap m_weight;
+};
+
+//
+// Vertex and Edge properties
+//
+#ifdef CSR
+typedef int weight_type;
+
+struct WeightedEdge {
+  WeightedEdge(weight_type weight = 0) : weight(weight) { }
+  
+  weight_type weight;
+};
+
+struct VertexProperties {
+  VertexProperties(int distance = 0, default_color_type color = white_color)
+    : distance(distance), color(color) { }
+
+  int distance;
+  default_color_type color;
+};
+#endif
+
+template <typename RandomGenerator, typename ProcessGroup, typename vertices_size_type,
+          typename edges_size_type>
+void
+run_non_distributed_graph_tests(RandomGenerator& gen, const ProcessGroup& pg,
+                                vertices_size_type n, edges_size_type m, 
+                                std::size_t maxEdgeWeight, uint64_t seed, 
+                                int K4Alpha, double a, double b, double c, double d, 
+                                int subGraphEdgeLength, bool show_degree_dist, 
+                                bool full_bc, bool verify)
+{
+#ifdef CSR
+  typedef compressed_sparse_row_graph<directedS, VertexProperties, WeightedEdge>
+    seqGraph;
+#else
+  typedef adjacency_list<vecS, vecS, directedS, 
+                         // Vertex properties
+                         property<vertex_distance_t, int,
+                         property<vertex_color_t, default_color_type> >,
+                         // Edge properties
+                         property<edge_weight_t, int> > seqGraph;
+#endif
+
+  // Generate sequential graph for non_distributed betweenness centrality
+  // Reseed the PRNG to get the same graph
+  gen.seed(seed);
+
+  synchronize(pg);
+
+  time_type start = get_time();
+
+#ifdef CSR
+  seqGraph sg(edges_are_sorted,
+              sorted_unique_rmat_iterator<RandomGenerator, seqGraph>(gen, n, m, a, b, c, d),
+              sorted_unique_rmat_iterator<RandomGenerator, seqGraph>(),
+              make_generator_iterator(gen, uniform_int<int>(0, maxEdgeWeight)),
+              n);
+#else
+  seqGraph sg(unique_rmat_iterator<RandomGenerator, seqGraph>(gen, n, m, a, b, c, d),
+              unique_rmat_iterator<RandomGenerator, seqGraph>(),
+              make_generator_iterator(gen, uniform_int<int>(0, maxEdgeWeight)),
+              n);
+#endif
+
+  // Not strictly necessary to synchronize here, but it make sure that the
+  // time we measure is the time needed for all copies of the graph to be 
+  // constructed
+  synchronize(pg);
+
+  time_type end = get_time();
+
+  if (process_id(pg) == 0)
+    std::cerr<< "Kernel 1:\n" 
+             << "    Non-Distributed Graph: " << print_time(end - start) << std::endl;
+
+  std::map<int, int> degree_dist;
+  if ( show_degree_dist ) {
+    BGL_FORALL_VERTICES_T(v, sg, seqGraph) {
+      degree_dist[out_degree(v, sg)]++;
+    }
+
+    std::cerr << "Degree - Fraction of vertices of that degree\n";
+    for (std::map<int, int>::iterator iter = degree_dist.begin(); 
+         iter != degree_dist.end(); ++iter)
+      std::cerr << "  " << iter->first << " - " << double(iter->second) / num_vertices(sg) << std::endl << std::endl;
+  }
+
+  //
+  // Kernel 2 - Classify large sets
+  //
+  std::vector<graph_traits<seqGraph>::edge_descriptor> seqS;
+
+  if (process_id(pg) == 0)
+    std::cerr << "Kernel 2:\n";
+
+  seq_classify_sets(pg, sg,
+#ifdef CSR
+                    get(&WeightedEdge::weight, sg),
+#else
+                    get(edge_weight, sg),
+#endif
+                    seqS);
+
+  //
+  // Kernel 3 - Graph Extraction
+  //
+#ifdef CSR
+  typedef weight_type weight_t;
+  weight_t unit_weight(1);
+#else
+  int unit_weight(1);;
+#endif
+
+  if (process_id(pg) == 0)
+    std::cerr << "Kernel 3:\n";
+
+  seq_subgraph_extraction(pg, sg,
+#ifdef CSR
+//                        get(&WeightedEdge::weight, sg), 
+                          ref_property_map<graph_traits<seqGraph>::edge_descriptor, weight_t>(unit_weight),
+                          get(&VertexProperties::distance, sg),
+                          get(&VertexProperties::color, sg), 
+#else
+//                        get(edge_weight, sg), 
+                          ref_property_map<graph_traits<seqGraph>::edge_descriptor, int>(unit_weight),
+                          get(vertex_distance, sg), 
+                          get(vertex_color, sg), 
+#endif
+                          seqS, subGraphEdgeLength); 
+
+#ifdef CSR
+  typedef property_map<seqGraph, weight_type WeightedEdge::*>::type seqEdgeWeightMap;
+  edge_weight_not_divisible_by_eight<seqEdgeWeightMap> sg_filter(get(&WeightedEdge::weight, sg));
+#else
+  typedef property_map<seqGraph, edge_weight_t>::type seqEdgeWeightMap;
+  edge_weight_not_divisible_by_eight<seqEdgeWeightMap> sg_filter(get(edge_weight, sg));
+#endif
+
+  typedef filtered_graph<const seqGraph, edge_weight_not_divisible_by_eight<seqEdgeWeightMap> > 
+    filteredSeqGraph;
+
+  filteredSeqGraph fsg(sg, sg_filter);
+
+  std::vector<graph_traits<seqGraph>::vertex_descriptor> max_seq_bc_vec;
+
+  // Non-Distributed Centrality Map
+  typedef property_map<seqGraph, vertex_index_t>::const_type seqIndexMap;
+  typedef iterator_property_map<std::vector<int>::iterator, seqIndexMap> seqCentralityMap;
+
+  std::vector<int> non_distributed_centralityS(num_vertices(sg), 0);
+  seqCentralityMap non_distributed_centrality(non_distributed_centralityS.begin(), 
+                                              get(vertex_index, sg));
+
+  vertices_size_type n0 = 0;
+  BGL_FORALL_VERTICES_T(v, fsg, filteredSeqGraph) {
+    if (out_degree(v, fsg) == 0) ++n0;
+  }
+
+  if (process_id(pg) == 0)
+    std::cerr << "Kernel 4:\n";
+
+  // Run Betweenness Centrality
+  if (full_bc) {
+
+    // Non-Distributed Graph BC
+    start = get_time();
+    non_distributed_brandes_betweenness_centrality(pg, fsg, non_distributed_centrality);
+    extract_max_bc_vertices(pg, fsg, non_distributed_centrality, max_seq_bc_vec);
+    end = get_time();
+
+    edges_size_type nonDistributedExactTEPs = edges_size_type(floor(7 * n* (n - n0) / (end - start)));
+
+    if (process_id(pg) == 0)
+      std::cerr << "    non-Distributed Graph Exact = " << print_time(end - start) << " ("
+                << nonDistributedExactTEPs << " TEPs)\n";
+  }
+
+  // Non-Distributed Graph Approximate BC
+  std::vector<int> nonDistributedApproxCentralityS(num_vertices(sg), 0);
+  seqCentralityMap nonDistributedApproxCentrality(nonDistributedApproxCentralityS.begin(), 
+                                                  get(vertex_index, sg));
+
+  queue<typename graph_traits<filteredSeqGraph>::vertex_descriptor> sources;
+  {
+    minstd_rand gen;
+    uniform_int<vertices_size_type> rand_vertex(0, num_vertices(fsg) - 1);
+    int remaining_sources = floor(pow(2, K4Alpha));
+    std::vector<typename graph_traits<filteredSeqGraph>::vertex_descriptor> temp_sources;
+ 
+    while (remaining_sources > 0) {
+      typename graph_traits<filteredSeqGraph>::vertex_descriptor v = 
+        vertex(rand_vertex(gen), fsg);
+    
+      if (out_degree(v, fsg) != 0
+          && std::find(temp_sources.begin(), temp_sources.end(), v) == temp_sources.end()) {
+        temp_sources.push_back(v);
+        --remaining_sources;
+      }
+    }
+
+    for (int i = 0; i < temp_sources.size(); ++i)
+      sources.push(temp_sources[i]);
+  }
+
+  start = get_time();
+  non_distributed_brandes_betweenness_centrality(pg, fsg, buffer(sources).
+                                                 centrality_map(nonDistributedApproxCentrality)); 
+  extract_max_bc_vertices(pg, fsg, nonDistributedApproxCentrality, max_seq_bc_vec);
+  end = get_time();
+
+  edges_size_type nonDistributedApproxTEPs = edges_size_type(floor(7 * n * pow(2, K4Alpha) / (end - start)));
+
+  if (process_id(pg) == 0)
+    std::cerr << "    Non-Distributed Graph Approximate (" << floor(pow(2, K4Alpha)) << " sources) = " 
+              << print_time(end - start) << " (" << nonDistributedApproxTEPs << " TEPs)\n";
+
+  // Verify Correctness of Kernel 4
+  if (full_bc && verify && process_id(pg) == 0) {
+
+    std::vector<int> seq_centralityS(num_vertices(fsg), 0);
+    seqCentralityMap seq_centrality(seq_centralityS.begin(), get(vertex_index, fsg));
+
+    max_seq_bc_vec.clear();
+    property_traits<seqCentralityMap>::value_type max_ = 0;
+
+    start = get_time();
+    brandes_betweenness_centrality(fsg, seq_centrality);
+
+    typedef filtered_graph<const seqGraph, edge_weight_not_divisible_by_eight<seqEdgeWeightMap> >
+      filteredSeqGraph;
+
+    BGL_FORALL_VERTICES_T(v, fsg, filteredSeqGraph ) {
+      if (get(seq_centrality, v) == max_)
+        max_seq_bc_vec.push_back(v);
+      else if (get(seq_centrality, v) > max_) {
+        max_ = get(seq_centrality, v);
+        max_seq_bc_vec.clear();
+        max_seq_bc_vec.push_back(v);      
+      }
+    }
+
+    end = get_time();
+
+    edges_size_type sequentialTEPs = edges_size_type(floor(7 * n* (n - n0) / (end - start)));
+
+    std::cerr << "    Sequential = " << print_time(end - start) << " (" << sequentialTEPs << " TEPs)\n";
+
+    typename ProcessGroup::process_id_type id = process_id(pg);
+    typename ProcessGroup::process_size_type p = num_processes(pg);    
+
+    assert((double)n/p == floor((double)n/p));
+    
+    std::cerr << "\nVerifying non-scalable betweenness centrality...\n";
+
+    {
+      bool passed = true;
+
+      // Verify non-scalable betweenness centrality
+      BGL_FORALL_VERTICES_T(v, sg, seqGraph) {
+        if (get(non_distributed_centrality, v) != get(seq_centrality, v)) {
+          std::cerr << "  " << id << ": Error - centrality of " << v
+                    << " does not match the sequential result (" 
+                    << get(non_distributed_centrality, v) << " vs. " 
+                    << get(seq_centrality, v) << ")\n";
+          passed = false;
+        }
+      }
+
+      if (passed)
+        std::cerr << "  PASSED\n";
+    }
+
+  }    
+
+}
+
+template <typename RandomGenerator, typename ProcessGroup, typename vertices_size_type,
+          typename edges_size_type>
+void
+run_distributed_graph_tests(RandomGenerator& gen, const ProcessGroup& pg,
+                            vertices_size_type n, edges_size_type m, 
+                            std::size_t maxEdgeWeight, uint64_t seed, 
+                            int K4Alpha, double a, double b, double c, double d, 
+                            int subGraphEdgeLength, bool show_degree_dist, 
+                            bool emit_dot_file, bool full_bc, bool verify)
+{
+#ifdef CSR
+  typedef compressed_sparse_row_graph<directedS, VertexProperties, WeightedEdge, no_property,
+                                      distributedS<ProcessGroup> > Graph;
+#else
+  typedef adjacency_list<vecS, 
+                         distributedS<ProcessGroup, vecS>,
+                         directedS, 
+                         // Vertex properties
+                         property<vertex_distance_t, int,
+                         property<vertex_color_t, default_color_type> >,
+                         // Edge properties
+                         property<edge_weight_t, int> > Graph;
+#endif
+
+  gen.seed(seed);
+
+  parallel::variant_distribution<ProcessGroup> distrib 
+    = parallel::block(pg, n);
+
+  typedef typename ProcessGroup::process_id_type process_id_type;
+  process_id_type id = process_id(pg);
+
+  typedef typename property_map<Graph, vertex_owner_t>::const_type OwnerMap;
+  typedef typename property_map<Graph, vertex_local_t>::const_type LocalMap;
+
+  typedef keep_local_edges<parallel::variant_distribution<ProcessGroup>,
+                           process_id_type>
+    EdgeFilter; 
+
+  //
+  // Kernel 1 - Graph construction
+  // Nick: The benchmark specifies that we only have to time graph generation from
+  //       edge tuples, the generator generates the edge tuples at graph construction
+  //       time so we're timing some overhead in the random number generator, etc.
+  synchronize(pg);
+
+  time_type start = get_time();
+
+#ifdef CSR
+//   typedef sorted_unique_rmat_iterator<RandomGenerator, Graph, EdgeFilter> RMATIter;
+  typedef sorted_rmat_iterator<RandomGenerator, Graph, keep_all_edges> RMATIter;
+
+  Graph g(//RMATIter(gen, n, m, a, b, c, d, false, true, EdgeFilter(distrib, id)),
+          RMATIter(gen, n, m, a, b, c, d, true, keep_all_edges()),
+          RMATIter(),
+          make_generator_iterator(gen, uniform_int<int>(0, maxEdgeWeight)),
+          n, pg, distrib);
+#else
+  typedef unique_rmat_iterator<RandomGenerator, Graph, EdgeFilter> RMATIter;
+  Graph g(RMATIter(gen, n, m, a, b, c, d, true EdgeFilter(distrib, id)),
+          RMATIter(),
+          make_generator_iterator(gen, uniform_int<int>(0, maxEdgeWeight)),
+          n, pg, distrib);
+#endif
+
+  synchronize(pg);
+
+  time_type end = get_time();
+
+  if (id == 0)
+    std::cerr<< "Kernel 1:\n" 
+             << "    Distributed Graph: " << print_time(end - start) << std::endl;
+
+  if ( emit_dot_file )
+    write_graphviz("ssca.dot", g);
+
+  //
+  // Kernel 2 - Classify large sets
+  //
+  typedef typename graph_traits<Graph>::vertex_descriptor vertex_descriptor;
+  std::vector<std::pair<vertex_descriptor, vertex_descriptor> > S;
+
+  if (id == 0)
+    std::cerr << "Kernel 2:\n";
+
+  classify_sets(g,
+#ifdef CSR
+                get(&WeightedEdge::weight, g),
+#else
+                get(edge_weight, g),
+#endif
+                S);
+
+  //
+  // Kernel 3 - Graph Extraction
+  //
+  OwnerMap owner = get(vertex_owner, g);
+  LocalMap local = get(vertex_local, g);
+
+  if (id == 0)
+    std::cerr << "Kernel 3:\n";
+
+#ifdef CSR
+  typedef weight_type weight_t;
+  weight_t unit_weight(1);
+#else
+  int unit_weight(1);;
+#endif
+
+  subgraph_extraction(g, owner, local, 
+#ifdef CSR
+//                    get(&WeightedEdge::weight, g), 
+                      ref_property_map<typename graph_traits<Graph>::edge_descriptor, weight_t>(unit_weight),
+                      get(&VertexProperties::distance, g),
+                      get(&VertexProperties::color, g),
+#else
+//                    get(edge_weight, g), 
+                      ref_property_map<graph_traits<Graph>::edge_descriptor, int>(unit_weight),
+                      get(vertex_distance, g), 
+                      get(vertex_color, g), 
+#endif
+                      S, subGraphEdgeLength);
+
+  //
+  // Kernel 4 - Betweenness Centrality
+  //
+
+  // Filter edges with weights divisible by 8
+#ifdef CSR
+  typedef typename property_map<Graph, weight_type WeightedEdge::*>::type EdgeWeightMap;
+  edge_weight_not_divisible_by_eight<EdgeWeightMap> filter(get(&WeightedEdge::weight, g));
+#else
+  typedef typename property_map<Graph, edge_weight_t>::type EdgeWeightMap;
+  edge_weight_not_divisible_by_eight<EdgeWeightMap> filter(get(edge_weight, g));
+#endif
+
+  typedef filtered_graph<const Graph, edge_weight_not_divisible_by_eight<EdgeWeightMap> > 
+    filteredGraph;
+  filteredGraph fg(g, filter);
+
+  // Vectors of max BC scores for all tests
+  std::vector<typename graph_traits<Graph>::vertex_descriptor> max_bc_vec; 
+
+  // Distributed Centrality Map
+  typedef typename property_map<Graph, vertex_index_t>::const_type IndexMap;
+  typedef iterator_property_map<std::vector<int>::iterator, IndexMap> CentralityMap;
+
+  std::vector<int> centralityS(num_vertices(g), 0);
+  CentralityMap centrality(centralityS.begin(), get(vertex_index, g));
+
+  // Calculate number of vertices of degree 0
+  vertices_size_type local_n0 = 0, n0;
+  BGL_FORALL_VERTICES_T(v, fg, filteredGraph) {
+    if (out_degree(v, g) == 0) local_n0++;
+  }
+  n0 = boost::parallel::all_reduce(pg, local_n0, std::plus<vertices_size_type>());
+
+  if (id == 0)
+    std::cerr << "Kernel 4:\n";
+
+  // Run Betweenness Centrality
+  if (full_bc) {
+    
+    // Distributed Graph Full BC
+    start = get_time();
+    brandes_betweenness_centrality(fg, centrality);
+    extract_max_bc_vertices(pg, g, centrality, max_bc_vec);
+    end = get_time();
+    
+    edges_size_type exactTEPs = edges_size_type(floor(7 * n* (n - n0) / (end - start)));
+    
+    if (id == 0)
+      std::cerr << "    Exact = " << print_time(end - start) << " ("
+                << exactTEPs << " TEPs)\n";
+  }
+
+  // Distributed Graph Approximate BC
+  std::vector<int> approxCentralityS(num_vertices(g), 0);
+  CentralityMap approxCentrality(approxCentralityS.begin(), get(vertex_index, g));
+
+  queue<vertex_descriptor> sources;
+  generate_sources(g, sources, vertices_size_type(floor(pow(2, K4Alpha))));
+  
+  start = get_time();
+  brandes_betweenness_centrality(fg, buffer(sources).centrality_map(approxCentrality)); 
+  extract_max_bc_vertices(pg, fg, approxCentrality, max_bc_vec);
+  end = get_time();
+  
+  edges_size_type approxTEPs = edges_size_type(floor(7 * n * pow(2, K4Alpha) / (end - start)));
+  
+  if (id == 0)
+    std::cerr << "    Approximate (" << floor(pow(2, K4Alpha)) << " sources) = " 
+              << print_time(end - start) << " (" << approxTEPs << " TEPs)\n";
+
+
+  // Verify Correctness of Kernel 4
+  if (full_bc && verify && id == 0) {
+
+    // Build non-distributed graph to verify against
+    typedef adjacency_list<vecS, vecS, directedS, 
+                           // Vertex properties
+                           property<vertex_distance_t, int,
+                           property<vertex_color_t, default_color_type> >,
+                           // Edge properties
+                           property<edge_weight_t, int> > seqGraph;
+
+    gen.seed(seed);
+
+#ifdef CSR
+    seqGraph sg(sorted_unique_rmat_iterator<RandomGenerator, seqGraph>(gen, n, m, a, b, c, d),
+                sorted_unique_rmat_iterator<RandomGenerator, seqGraph>(),
+                make_generator_iterator(gen, uniform_int<int>(0, maxEdgeWeight)),
+                n);
+#else
+    seqGraph sg(unique_rmat_iterator<RandomGenerator, seqGraph>(gen, n, m, a, b, c, d),
+                unique_rmat_iterator<RandomGenerator, seqGraph>(),
+                make_generator_iterator(gen, uniform_int<int>(0, maxEdgeWeight)),
+                n);
+#endif
+
+    typedef property_map<seqGraph, edge_weight_t>::type seqEdgeWeightMap;
+    edge_weight_not_divisible_by_eight<seqEdgeWeightMap> sg_filter(get(edge_weight, sg));
+    
+    filtered_graph<const seqGraph, edge_weight_not_divisible_by_eight<seqEdgeWeightMap> > 
+      fsg(sg, sg_filter);
+
+    // Build sequential centrality map 
+    typedef property_map<seqGraph, vertex_index_t>::const_type seqIndexMap;
+    typedef iterator_property_map<std::vector<int>::iterator, seqIndexMap> seqCentralityMap;
+
+    std::vector<int> seq_centralityS(num_vertices(sg), 0);
+    seqCentralityMap seq_centrality(seq_centralityS.begin(), get(vertex_index, sg));
+
+    std::vector<graph_traits<seqGraph>::vertex_descriptor> max_seq_bc_vec;
+
+    max_seq_bc_vec.clear();
+    property_traits<seqCentralityMap>::value_type max_ = 0;
+
+    start = get_time();
+    brandes_betweenness_centrality(fsg, seq_centrality);
+
+    typedef filtered_graph<const seqGraph, edge_weight_not_divisible_by_eight<seqEdgeWeightMap> >
+      filteredSeqGraph;
+
+    BGL_FORALL_VERTICES_T(v, fsg, filteredSeqGraph ) {
+      if (get(seq_centrality, v) == max_)
+        max_seq_bc_vec.push_back(v);
+      else if (get(seq_centrality, v) > max_) {
+        max_ = get(seq_centrality, v);
+        max_seq_bc_vec.clear();
+        max_seq_bc_vec.push_back(v);      
+      }
+    }
+
+    end = get_time();
+
+    edges_size_type sequentialTEPs = edges_size_type(floor(7 * n* (n - n0) / (end - start)));
+
+    std::cerr << "    Sequential = " << print_time(end - start) << " (" << sequentialTEPs << " TEPs)\n";
+    
+    typename ProcessGroup::process_size_type p = num_processes(pg);    
+
+    assert((double)n/p == floor((double)n/p));
+
+    std::cerr << "\nVerifying betweenness centrality...\n";
+
+    {
+      bool passed = true;
+
+      // Verify exact betweenness centrality
+      BGL_FORALL_VERTICES_T(v, g, Graph) {
+        if (get(centrality, v) != seq_centralityS[(n/p) * get(owner, v) + get(local, v)]) {
+          std::cerr << "  " << id << ": Error - centrality of " << get(local, v) << "@" << get(owner, v)
+                    << " does not match the sequential result (" << get(centrality, v) << " vs. " 
+                    << seq_centralityS[(n/p) * get(owner, v) + get(local, v)] << ")\n";
+          passed = false;
+        }
+      }
+
+      if (passed)
+        std::cerr << "  PASSED\n";
+    }      
+  }    
+}
+
+void usage()
+{
+  std::cerr << "SSCA benchmark.\n\n"
+            << "Usage : ssca [options]\n\n"
+            << "Options are:\n"
+            << "\t--vertices v\t\t\tNumber of vertices in the graph\n"
+            << "\t--edges v\t\t\tNumber of edges in the graph\n"
+            << "\t--seed s\t\t\tSeed for synchronized random number generator\n"
+            << "\t--full-bc\t\t\tRun full (exact) Betweenness Centrality\n"
+            << "\t--max-weight miw\t\tMaximum integer edge weight\n"
+            << "\t--subgraph-edge-length sel\tEdge length of subgraphs to extract in Kernel 3\n"
+            << "\t--k4alpha k\t\t\tValue of K4Alpha in Kernel 4\n"
+            << "\t--scale s\t\t\tSCALE parameter for the SSCA benchmark (sets n, m, and C)\n"
+            << "\t--dot\t\t\t\tEmit a dot file containing the graph\n"
+            << "\t--verify\t\t\tVerify result\n"
+            << "\t--degree-dist\t\t\t Output degree distribution of graph\n"
+            << "\t--no-distributed-graph\t\tOmit distributed graph tests\n";
+}
+
+int test_main(int argc, char* argv[])
+{
+  mpi::environment env(argc, argv);
+
+  using boost::graph::distributed::mpi_process_group;
+
+#ifdef CSR
+  typedef compressed_sparse_row_graph<directedS, VertexProperties, WeightedEdge, no_property,
+                                      distributedS<mpi_process_group> > Graph;
+#else
+  typedef adjacency_list<vecS, 
+                         distributedS<mpi_process_group, vecS>,
+                         directedS, 
+                         // Vertex properties
+                         property<vertex_distance_t, int,
+                         property<vertex_color_t, default_color_type> >,
+                         // Edge properties
+                         property<edge_weight_t, int> > Graph;
+#endif
+
+  typedef graph_traits<Graph>::vertices_size_type vertices_size_type;
+  typedef graph_traits<Graph>::edges_size_type edges_size_type;
+
+  RandomGenerator gen;
+  
+  // Default args
+  vertices_size_type n = 100;
+  edges_size_type m = 8*n;
+  uint64_t seed = 1;
+  int maxEdgeWeight = 100, 
+      subGraphEdgeLength = 8,
+      K4Alpha = 0.5;
+  double a = 0.57, b = 0.19, c = 0.19, d = 0.05;
+  bool emit_dot_file = false, verify = false, full_bc = true,
+    distributed_graph = true, show_degree_dist = false,
+    non_distributed_graph = true;
+
+  mpi_process_group pg;
+
+  if (argc == 1) {
+    if (process_id(pg) == 0)
+      usage();
+    exit(-1);
+  }
+
+  // Parse args
+  for (int i = 1; i < argc; ++i) {
+    std::string arg = argv[i];
+
+    if (arg == "--vertices")
+      n = boost::lexical_cast<vertices_size_type>( argv[i+1] );
+
+    if (arg == "--seed")
+      seed = boost::lexical_cast<uint64_t>( argv[i+1] );
+
+    if (arg == "--full-bc")
+      full_bc = (argv[i+1]== "true");
+
+    if (arg == "--max-weight")
+      maxEdgeWeight = boost::lexical_cast<int>( argv[i+1] );
+
+    if (arg == "--subgraph-edge-length")
+      subGraphEdgeLength = boost::lexical_cast<int>( argv[i+1] );
+    
+    if (arg == "--edges")
+      m = boost::lexical_cast<edges_size_type>( argv[i+1] );
+
+    if (arg == "--k4alpha")
+      K4Alpha = boost::lexical_cast<int>( argv[i+1] );
+
+    if (arg == "--dot")
+      emit_dot_file = true;
+
+    if (arg == "--verify")
+      verify = true;
+
+    if (arg == "--degree-dist")
+      show_degree_dist = true;
+
+    if (arg == "--no-distributed-graph")
+      distributed_graph = false;
+
+    if (arg == "--no-non-distributed-graph")
+      non_distributed_graph = false;
+
+    if (arg == "--scale") {
+      vertices_size_type scale  = boost::lexical_cast<vertices_size_type>( argv[i+1] );
+      maxEdgeWeight = n = vertices_size_type(floor(pow(2, scale)));
+      m = 8 * n;
+    }
+
+    if (arg == "--help") {
+      if (process_id(pg) == 0)
+        usage();
+      exit(-1);
+    }
+  }
+
+  if (non_distributed_graph) {
+    if (process_id(pg) == 0)
+      std::cerr << "Non-Distributed Graph Tests\n";
+
+    run_non_distributed_graph_tests(gen, pg, n, m, maxEdgeWeight, seed, K4Alpha, a, b, c, d, 
+                                    subGraphEdgeLength, show_degree_dist, full_bc, verify);
+  }
+
+  if (distributed_graph) {
+    if (process_id(pg) == 0)
+      std::cerr << "Distributed Graph Tests\n";
+
+    run_distributed_graph_tests(gen, pg, n, m, maxEdgeWeight, seed, K4Alpha, a, b, c, d, 
+                                subGraphEdgeLength, show_degree_dist, emit_dot_file, 
+                                full_bc, verify);
+  }
+
+  return 0;
+}