path: root/doc/src/sgml/html/indexes-ordering.html
diff options
Diffstat (limited to 'doc/src/sgml/html/indexes-ordering.html')
1 files changed, 64 insertions, 0 deletions
diff --git a/doc/src/sgml/html/indexes-ordering.html b/doc/src/sgml/html/indexes-ordering.html
new file mode 100644
index 0000000..658cb51
--- /dev/null
+++ b/doc/src/sgml/html/indexes-ordering.html
@@ -0,0 +1,64 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" ""><html xmlns=""><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /><title>11.4. Indexes and ORDER BY</title><link rel="stylesheet" type="text/css" href="stylesheet.css" /><link rev="made" href="" /><meta name="generator" content="DocBook XSL Stylesheets V1.79.1" /><link rel="prev" href="indexes-multicolumn.html" title="11.3. Multicolumn Indexes" /><link rel="next" href="indexes-bitmap-scans.html" title="11.5. Combining Multiple Indexes" /></head><body><div xmlns="" class="navheader"><table width="100%" summary="Navigation header"><tr><th colspan="5" align="center">11.4. Indexes and <code xmlns="" class="literal">ORDER BY</code></th></tr><tr><td width="10%" align="left"><a accesskey="p" href="indexes-multicolumn.html" title="11.3. Multicolumn Indexes">Prev</a> </td><td width="10%" align="left"><a accesskey="u" href="indexes.html" title="Chapter 11. Indexes">Up</a></td><th width="60%" align="center">Chapter 11. Indexes</th><td width="10%" align="right"><a accesskey="h" href="index.html" title="PostgreSQL 12.0 Documentation">Home</a></td><td width="10%" align="right"> <a accesskey="n" href="indexes-bitmap-scans.html" title="11.5. Combining Multiple Indexes">Next</a></td></tr></table><hr></hr></div><div class="sect1" id="INDEXES-ORDERING"><div class="titlepage"><div><div><h2 class="title" style="clear: both">11.4. Indexes and <code class="literal">ORDER BY</code></h2></div></div></div><a id="id-" class="indexterm"></a><p>
+ In addition to simply finding the rows to be returned by a query,
+ an index may be able to deliver them in a specific sorted order.
+ This allows a query's <code class="literal">ORDER BY</code> specification to be honored
+ without a separate sorting step. Of the index types currently
+ supported by <span class="productname">PostgreSQL</span>, only B-tree
+ can produce sorted output — the other index types return
+ matching rows in an unspecified, implementation-dependent order.
+ </p><p>
+ The planner will consider satisfying an <code class="literal">ORDER BY</code> specification
+ either by scanning an available index that matches the specification,
+ or by scanning the table in physical order and doing an explicit
+ sort. For a query that requires scanning a large fraction of the
+ table, an explicit sort is likely to be faster than using an index
+ because it requires
+ less disk I/O due to following a sequential access pattern. Indexes are
+ more useful when only a few rows need be fetched. An important
+ special case is <code class="literal">ORDER BY</code> in combination with
+ <code class="literal">LIMIT</code> <em class="replaceable"><code>n</code></em>: an explicit sort will have to process
+ all the data to identify the first <em class="replaceable"><code>n</code></em> rows, but if there is
+ an index matching the <code class="literal">ORDER BY</code>, the first <em class="replaceable"><code>n</code></em>
+ rows can be retrieved directly, without scanning the remainder at all.
+ </p><p>
+ By default, B-tree indexes store their entries in ascending order
+ with nulls last (table TID is treated as a tiebreaker column among
+ otherwise equal entries). This means that a forward scan of an
+ index on column <code class="literal">x</code> produces output satisfying <code class="literal">ORDER BY x</code>
+ (or more verbosely, <code class="literal">ORDER BY x ASC NULLS LAST</code>). The
+ index can also be scanned backward, producing output satisfying
+ <code class="literal">ORDER BY x DESC</code>
+ (or more verbosely, <code class="literal">ORDER BY x DESC NULLS FIRST</code>, since
+ <code class="literal">NULLS FIRST</code> is the default for <code class="literal">ORDER BY DESC</code>).
+ </p><p>
+ You can adjust the ordering of a B-tree index by including the
+ options <code class="literal">ASC</code>, <code class="literal">DESC</code>, <code class="literal">NULLS FIRST</code>,
+ and/or <code class="literal">NULLS LAST</code> when creating the index; for example:
+</p><pre class="programlisting">
+CREATE INDEX test2_info_nulls_low ON test2 (info NULLS FIRST);
+CREATE INDEX test3_desc_index ON test3 (id DESC NULLS LAST);
+ An index stored in ascending order with nulls first can satisfy
+ either <code class="literal">ORDER BY x ASC NULLS FIRST</code> or
+ <code class="literal">ORDER BY x DESC NULLS LAST</code> depending on which direction
+ it is scanned in.
+ </p><p>
+ You might wonder why bother providing all four options, when two
+ options together with the possibility of backward scan would cover
+ all the variants of <code class="literal">ORDER BY</code>. In single-column indexes
+ the options are indeed redundant, but in multicolumn indexes they can be
+ useful. Consider a two-column index on <code class="literal">(x, y)</code>: this can
+ satisfy <code class="literal">ORDER BY x, y</code> if we scan forward, or
+ <code class="literal">ORDER BY x DESC, y DESC</code> if we scan backward.
+ But it might be that the application frequently needs to use
+ <code class="literal">ORDER BY x ASC, y DESC</code>. There is no way to get that
+ ordering from a plain index, but it is possible if the index is defined
+ as <code class="literal">(x ASC, y DESC)</code> or <code class="literal">(x DESC, y ASC)</code>.
+ </p><p>
+ Obviously, indexes with non-default sort orderings are a fairly
+ specialized feature, but sometimes they can produce tremendous
+ speedups for certain queries. Whether it's worth maintaining such an
+ index depends on how often you use queries that require a special
+ sort ordering.
+ </p></div><div class="navfooter"><hr /><table width="100%" summary="Navigation footer"><tr><td width="40%" align="left"><a accesskey="p" href="indexes-multicolumn.html">Prev</a> </td><td width="20%" align="center"><a accesskey="u" href="indexes.html">Up</a></td><td width="40%" align="right"> <a accesskey="n" href="indexes-bitmap-scans.html">Next</a></td></tr><tr><td width="40%" align="left" valign="top">11.3. Multicolumn Indexes </td><td width="20%" align="center"><a accesskey="h" href="index.html">Home</a></td><td width="40%" align="right" valign="top"> 11.5. Combining Multiple Indexes</td></tr></table></div></body></html> \ No newline at end of file