diff options
Diffstat (limited to 'doc/src/sgml/html/different-replication-solutions.html')
| -rw-r--r-- | doc/src/sgml/html/different-replication-solutions.html | 137 |
1 files changed, 137 insertions, 0 deletions
diff --git a/doc/src/sgml/html/different-replication-solutions.html b/doc/src/sgml/html/different-replication-solutions.html new file mode 100644 index 0000000..4919117 --- /dev/null +++ b/doc/src/sgml/html/different-replication-solutions.html @@ -0,0 +1,137 @@ +<?xml version="1.0" encoding="UTF-8" standalone="no"?> +<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><html xmlns="http://www.w3.org/1999/xhtml"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /><title>27.1. Comparison of Different Solutions</title><link rel="stylesheet" type="text/css" href="stylesheet.css" /><link rev="made" href="pgsql-docs@lists.postgresql.org" /><meta name="generator" content="DocBook XSL Stylesheets Vsnapshot" /><link rel="prev" href="high-availability.html" title="Chapter 27. High Availability, Load Balancing, and Replication" /><link rel="next" href="warm-standby.html" title="27.2. Log-Shipping Standby Servers" /></head><body id="docContent" class="container-fluid col-10"><div class="navheader"><table width="100%" summary="Navigation header"><tr><th colspan="5" align="center">27.1. Comparison of Different Solutions</th></tr><tr><td width="10%" align="left"><a accesskey="p" href="high-availability.html" title="Chapter 27. High Availability, Load Balancing, and Replication">Prev</a> </td><td width="10%" align="left"><a accesskey="u" href="high-availability.html" title="Chapter 27. High Availability, Load Balancing, and Replication">Up</a></td><th width="60%" align="center">Chapter 27. High Availability, Load Balancing, and Replication</th><td width="10%" align="right"><a accesskey="h" href="index.html" title="PostgreSQL 15.5 Documentation">Home</a></td><td width="10%" align="right"> <a accesskey="n" href="warm-standby.html" title="27.2. Log-Shipping Standby Servers">Next</a></td></tr></table><hr /></div><div class="sect1" id="DIFFERENT-REPLICATION-SOLUTIONS"><div class="titlepage"><div><div><h2 class="title" style="clear: both">27.1. Comparison of Different Solutions</h2></div></div></div><div class="variablelist"><dl class="variablelist"><dt><span class="term">Shared Disk Failover</span></dt><dd><p> + Shared disk failover avoids synchronization overhead by having only one + copy of the database. It uses a single disk array that is shared by + multiple servers. If the main database server fails, the standby server + is able to mount and start the database as though it were recovering from + a database crash. This allows rapid failover with no data loss. + </p><p> + Shared hardware functionality is common in network storage devices. + Using a network file system is also possible, though care must be + taken that the file system has full <acronym class="acronym">POSIX</acronym> behavior (see <a class="xref" href="creating-cluster.html#CREATING-CLUSTER-NFS" title="19.2.2.1. NFS">Section 19.2.2.1</a>). One significant limitation of this + method is that if the shared disk array fails or becomes corrupt, the + primary and standby servers are both nonfunctional. Another issue is + that the standby server should never access the shared storage while + the primary server is running. + </p></dd><dt><span class="term">File System (Block Device) Replication</span></dt><dd><p> + A modified version of shared hardware functionality is file system + replication, where all changes to a file system are mirrored to a file + system residing on another computer. The only restriction is that + the mirroring must be done in a way that ensures the standby server + has a consistent copy of the file system — specifically, writes + to the standby must be done in the same order as those on the primary. + <span class="productname">DRBD</span> is a popular file system replication solution + for Linux. + </p></dd><dt><span class="term">Write-Ahead Log Shipping</span></dt><dd><p> + Warm and hot standby servers can be kept current by reading a + stream of write-ahead log (<acronym class="acronym">WAL</acronym>) + records. If the main server fails, the standby contains + almost all of the data of the main server, and can be quickly + made the new primary database server. This can be synchronous or + asynchronous and can only be done for the entire database server. + </p><p> + A standby server can be implemented using file-based log shipping + (<a class="xref" href="warm-standby.html" title="27.2. Log-Shipping Standby Servers">Section 27.2</a>) or streaming replication (see + <a class="xref" href="warm-standby.html#STREAMING-REPLICATION" title="27.2.5. Streaming Replication">Section 27.2.5</a>), or a combination of both. For + information on hot standby, see <a class="xref" href="hot-standby.html" title="27.4. Hot Standby">Section 27.4</a>. + </p></dd><dt><span class="term">Logical Replication</span></dt><dd><p> + Logical replication allows a database server to send a stream of data + modifications to another server. <span class="productname">PostgreSQL</span> + logical replication constructs a stream of logical data modifications + from the WAL. Logical replication allows replication of data changes on + a per-table basis. In addition, a server that is publishing its own + changes can also subscribe to changes from another server, allowing data + to flow in multiple directions. For more information on logical + replication, see <a class="xref" href="logical-replication.html" title="Chapter 31. Logical Replication">Chapter 31</a>. Through the + logical decoding interface (<a class="xref" href="logicaldecoding.html" title="Chapter 49. Logical Decoding">Chapter 49</a>), + third-party extensions can also provide similar functionality. + </p></dd><dt><span class="term">Trigger-Based Primary-Standby Replication</span></dt><dd><p> + A trigger-based replication setup typically funnels data modification + queries to a designated primary server. Operating on a per-table basis, + the primary server sends data changes (typically) asynchronously to the + standby servers. Standby servers can answer queries while the primary is + running, and may allow some local data changes or write activity. This + form of replication is often used for offloading large analytical or data + warehouse queries. + </p><p> + <span class="productname">Slony-I</span> is an example of this type of + replication, with per-table granularity, and support for multiple standby + servers. Because it updates the standby server asynchronously (in + batches), there is possible data loss during fail over. + </p></dd><dt><span class="term">SQL-Based Replication Middleware</span></dt><dd><p> + With SQL-based replication middleware, a program intercepts + every SQL query and sends it to one or all servers. Each server + operates independently. Read-write queries must be sent to all servers, + so that every server receives any changes. But read-only queries can be + sent to just one server, allowing the read workload to be distributed + among them. + </p><p> + If queries are simply broadcast unmodified, functions like + <code class="function">random()</code>, <code class="function">CURRENT_TIMESTAMP</code>, and + sequences can have different values on different servers. + This is because each server operates independently, and because + SQL queries are broadcast rather than actual data changes. If + this is unacceptable, either the middleware or the application + must determine such values from a single source and then use those + values in write queries. Care must also be taken that all + transactions either commit or abort on all servers, perhaps + using two-phase commit (<a class="xref" href="sql-prepare-transaction.html" title="PREPARE TRANSACTION"><span class="refentrytitle">PREPARE TRANSACTION</span></a> + and <a class="xref" href="sql-commit-prepared.html" title="COMMIT PREPARED"><span class="refentrytitle">COMMIT PREPARED</span></a>). + <span class="productname">Pgpool-II</span> and <span class="productname">Continuent Tungsten</span> + are examples of this type of replication. + </p></dd><dt><span class="term">Asynchronous Multimaster Replication</span></dt><dd><p> + For servers that are not regularly connected or have slow + communication links, like laptops or + remote servers, keeping data consistent among servers is a + challenge. Using asynchronous multimaster replication, each + server works independently, and periodically communicates with + the other servers to identify conflicting transactions. The + conflicts can be resolved by users or conflict resolution rules. + Bucardo is an example of this type of replication. + </p></dd><dt><span class="term">Synchronous Multimaster Replication</span></dt><dd><p> + In synchronous multimaster replication, each server can accept + write requests, and modified data is transmitted from the + original server to every other server before each transaction + commits. Heavy write activity can cause excessive locking and + commit delays, leading to poor performance. Read requests can + be sent to any server. Some implementations use shared disk + to reduce the communication overhead. Synchronous multimaster + replication is best for mostly read workloads, though its big + advantage is that any server can accept write requests — + there is no need to partition workloads between primary and + standby servers, and because the data changes are sent from one + server to another, there is no problem with non-deterministic + functions like <code class="function">random()</code>. + </p><p> + <span class="productname">PostgreSQL</span> does not offer this type of replication, + though <span class="productname">PostgreSQL</span> two-phase commit (<a class="xref" href="sql-prepare-transaction.html" title="PREPARE TRANSACTION"><span class="refentrytitle">PREPARE TRANSACTION</span></a> and <a class="xref" href="sql-commit-prepared.html" title="COMMIT PREPARED"><span class="refentrytitle">COMMIT PREPARED</span></a>) + can be used to implement this in application code or middleware. + </p></dd></dl></div><p> + <a class="xref" href="different-replication-solutions.html#HIGH-AVAILABILITY-MATRIX" title="Table 27.1. High Availability, Load Balancing, and Replication Feature Matrix">Table 27.1</a> summarizes + the capabilities of the various solutions listed above. + </p><div class="table" id="HIGH-AVAILABILITY-MATRIX"><p class="title"><strong>Table 27.1. High Availability, Load Balancing, and Replication Feature Matrix</strong></p><div class="table-contents"><table class="table" summary="High Availability, Load Balancing, and Replication Feature Matrix" border="1"><colgroup><col class="col1" /><col class="col2" /><col class="col3" /><col class="col4" /><col class="col5" /><col class="col6" /><col class="col7" /><col class="col8" /><col class="col9" /></colgroup><thead><tr><th>Feature</th><th>Shared Disk</th><th>File System Repl.</th><th>Write-Ahead Log Shipping</th><th>Logical Repl.</th><th>Trigger-Based Repl.</th><th>SQL Repl. Middle-ware</th><th>Async. MM Repl.</th><th>Sync. MM Repl.</th></tr></thead><tbody><tr><td>Popular examples</td><td align="center">NAS</td><td align="center">DRBD</td><td align="center">built-in streaming repl.</td><td align="center">built-in logical repl., pglogical</td><td align="center">Londiste, Slony</td><td align="center">pgpool-II</td><td align="center">Bucardo</td><td align="center"> </td></tr><tr><td>Comm. method</td><td align="center">shared disk</td><td align="center">disk blocks</td><td align="center">WAL</td><td align="center">logical decoding</td><td align="center">table rows</td><td align="center">SQL</td><td align="center">table rows</td><td align="center">table rows and row locks</td></tr><tr><td>No special hardware required</td><td align="center"> </td><td align="center">•</td><td align="center">•</td><td align="center">•</td><td align="center">•</td><td align="center">•</td><td align="center">•</td><td align="center">•</td></tr><tr><td>Allows multiple primary servers</td><td align="center"> </td><td align="center"> </td><td align="center"> </td><td align="center">•</td><td align="center"> </td><td align="center">•</td><td align="center">•</td><td align="center">•</td></tr><tr><td>No overhead on primary</td><td align="center">•</td><td align="center"> </td><td align="center">•</td><td align="center">•</td><td align="center"> </td><td align="center">•</td><td align="center"> </td><td align="center"> </td></tr><tr><td>No waiting for multiple servers</td><td align="center">•</td><td align="center"> </td><td align="center">with sync off</td><td align="center">with sync off</td><td align="center">•</td><td align="center"> </td><td align="center">•</td><td align="center"> </td></tr><tr><td>Primary failure will never lose data</td><td align="center">•</td><td align="center">•</td><td align="center">with sync on</td><td align="center">with sync on</td><td align="center"> </td><td align="center">•</td><td align="center"> </td><td align="center">•</td></tr><tr><td>Replicas accept read-only queries</td><td align="center"> </td><td align="center"> </td><td align="center">with hot standby</td><td align="center">•</td><td align="center">•</td><td align="center">•</td><td align="center">•</td><td align="center">•</td></tr><tr><td>Per-table granularity</td><td align="center"> </td><td align="center"> </td><td align="center"> </td><td align="center">•</td><td align="center">•</td><td align="center"> </td><td align="center">•</td><td align="center">•</td></tr><tr><td>No conflict resolution necessary</td><td align="center">•</td><td align="center">•</td><td align="center">•</td><td align="center"> </td><td align="center">•</td><td align="center">•</td><td align="center"> </td><td align="center">•</td></tr></tbody></table></div></div><br class="table-break" /><p> + There are a few solutions that do not fit into the above categories: + </p><div class="variablelist"><dl class="variablelist"><dt><span class="term">Data Partitioning</span></dt><dd><p> + Data partitioning splits tables into data sets. Each set can + be modified by only one server. For example, data can be + partitioned by offices, e.g., London and Paris, with a server + in each office. If queries combining London and Paris data + are necessary, an application can query both servers, or + primary/standby replication can be used to keep a read-only copy + of the other office's data on each server. + </p></dd><dt><span class="term">Multiple-Server Parallel Query Execution</span></dt><dd><p> + Many of the above solutions allow multiple servers to handle multiple + queries, but none allow a single query to use multiple servers to + complete faster. This solution allows multiple servers to work + concurrently on a single query. It is usually accomplished by + splitting the data among servers and having each server execute its + part of the query and return results to a central server where they + are combined and returned to the user. This can be implemented using the + <span class="productname">PL/Proxy</span> tool set. + </p></dd></dl></div><p> + It should also be noted that because <span class="productname">PostgreSQL</span> + is open source and easily extended, a number of companies have + taken <span class="productname">PostgreSQL</span> and created commercial + closed-source solutions with unique failover, replication, and load + balancing capabilities. These are not discussed here. + </p></div><div class="navfooter"><hr /><table width="100%" summary="Navigation footer"><tr><td width="40%" align="left"><a accesskey="p" href="high-availability.html" title="Chapter 27. High Availability, Load Balancing, and Replication">Prev</a> </td><td width="20%" align="center"><a accesskey="u" href="high-availability.html" title="Chapter 27. High Availability, Load Balancing, and Replication">Up</a></td><td width="40%" align="right"> <a accesskey="n" href="warm-standby.html" title="27.2. Log-Shipping Standby Servers">Next</a></td></tr><tr><td width="40%" align="left" valign="top">Chapter 27. High Availability, Load Balancing, and Replication </td><td width="20%" align="center"><a accesskey="h" href="index.html" title="PostgreSQL 15.5 Documentation">Home</a></td><td width="40%" align="right" valign="top"> 27.2. Log-Shipping Standby Servers</td></tr></table></div></body></html>
\ No newline at end of file |