<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<HTML
><HEAD
><TITLE
>Statistics Used by the Planner</TITLE
><META
NAME="GENERATOR"
CONTENT="Modular DocBook HTML Stylesheet Version 1.79"><LINK
REV="MADE"
HREF="mailto:pgsql-docs@postgresql.org"><LINK
REL="HOME"
TITLE="PostgreSQL 8.3.6 Documentation"
HREF="index.html"><LINK
REL="UP"
TITLE="Performance Tips"
HREF="performance-tips.html"><LINK
REL="PREVIOUS"
TITLE="Using EXPLAIN"
HREF="using-explain.html"><LINK
REL="NEXT"
TITLE="Controlling the Planner with Explicit JOIN Clauses"
HREF="explicit-joins.html"><LINK
REL="STYLESHEET"
TYPE="text/css"
HREF="stylesheet.css"><META
HTTP-EQUIV="Content-Type"
CONTENT="text/html; charset=ISO-8859-1"><META
NAME="creation"
CONTENT="2009-02-03T04:34:16"></HEAD
><BODY
CLASS="SECT1"
><DIV
CLASS="NAVHEADER"
><TABLE
SUMMARY="Header navigation table"
WIDTH="100%"
BORDER="0"
CELLPADDING="0"
CELLSPACING="0"
><TR
><TH
COLSPAN="5"
ALIGN="center"
VALIGN="bottom"
>PostgreSQL 8.3.6 Documentation</TH
></TR
><TR
><TD
WIDTH="10%"
ALIGN="left"
VALIGN="top"
><A
HREF="using-explain.html"
ACCESSKEY="P"
>Prev</A
></TD
><TD
WIDTH="10%"
ALIGN="left"
VALIGN="top"
><A
HREF="performance-tips.html"
>Fast Backward</A
></TD
><TD
WIDTH="60%"
ALIGN="center"
VALIGN="bottom"
>Chapter 14. Performance Tips</TD
><TD
WIDTH="10%"
ALIGN="right"
VALIGN="top"
><A
HREF="performance-tips.html"
>Fast Forward</A
></TD
><TD
WIDTH="10%"
ALIGN="right"
VALIGN="top"
><A
HREF="explicit-joins.html"
ACCESSKEY="N"
>Next</A
></TD
></TR
></TABLE
><HR
ALIGN="LEFT"
WIDTH="100%"></DIV
><DIV
CLASS="SECT1"
><H1
CLASS="SECT1"
><A
NAME="PLANNER-STATS"
>14.2. Statistics Used by the Planner</A
></H1
><A
NAME="AEN20129"
></A
><P
> As we saw in the previous section, the query planner needs to estimate
the number of rows retrieved by a query in order to make good choices
of query plans. This section provides a quick look at the statistics
that the system uses for these estimates.
</P
><P
> One component of the statistics is the total number of entries in
each table and index, as well as the number of disk blocks occupied
by each table and index. This information is kept in the table
<A
HREF="catalog-pg-class.html"
><TT
CLASS="STRUCTNAME"
>pg_class</TT
></A
>,
in the columns <TT
CLASS="STRUCTFIELD"
>reltuples</TT
> and
<TT
CLASS="STRUCTFIELD"
>relpages</TT
>. We can look at it with
queries similar to this one:
</P><PRE
CLASS="SCREEN"
>SELECT relname, relkind, reltuples, relpages FROM pg_class WHERE relname LIKE 'tenk1%';
relname | relkind | reltuples | relpages
----------------------+---------+-----------+----------
tenk1 | r | 10000 | 358
tenk1_hundred | i | 10000 | 30
tenk1_thous_tenthous | i | 10000 | 30
tenk1_unique1 | i | 10000 | 30
tenk1_unique2 | i | 10000 | 30
(5 rows)</PRE
><P>
Here we can see that <TT
CLASS="STRUCTNAME"
>tenk1</TT
> contains 10000
rows, as do its indexes, but the indexes are (unsurprisingly) much
smaller than the table.
</P
><P
> For efficiency reasons, <TT
CLASS="STRUCTFIELD"
>reltuples</TT
>
and <TT
CLASS="STRUCTFIELD"
>relpages</TT
> are not updated on-the-fly,
and so they usually contain somewhat out-of-date values.
They are updated by <TT
CLASS="COMMAND"
>VACUUM</TT
>, <TT
CLASS="COMMAND"
>ANALYZE</TT
>, and a
few DDL commands such as <TT
CLASS="COMMAND"
>CREATE INDEX</TT
>. A stand-alone
<TT
CLASS="COMMAND"
>ANALYZE</TT
>, that is one not part of <TT
CLASS="COMMAND"
>VACUUM</TT
>,
generates an approximate <TT
CLASS="STRUCTFIELD"
>reltuples</TT
> value
since it does not read every row of the table. The planner
will scale the values it finds in <TT
CLASS="STRUCTNAME"
>pg_class</TT
>
to match the current physical table size, thus obtaining a closer
approximation.
</P
><A
NAME="AEN20150"
></A
><P
> Most queries retrieve only a fraction of the rows in a table, due
to having <TT
CLASS="LITERAL"
>WHERE</TT
> clauses that restrict the rows to be
examined. The planner thus needs to make an estimate of the
<I
CLASS="FIRSTTERM"
>selectivity</I
> of <TT
CLASS="LITERAL"
>WHERE</TT
> clauses, that is,
the fraction of rows that match each condition in the
<TT
CLASS="LITERAL"
>WHERE</TT
> clause. The information used for this task is
stored in the
<A
HREF="catalog-pg-statistic.html"
><TT
CLASS="STRUCTNAME"
>pg_statistic</TT
></A
>
system catalog. Entries in <TT
CLASS="STRUCTNAME"
>pg_statistic</TT
>
are updated by the <TT
CLASS="COMMAND"
>ANALYZE</TT
> and <TT
CLASS="COMMAND"
>VACUUM
ANALYZE</TT
> commands, and are always approximate even when freshly
updated.
</P
><A
NAME="AEN20162"
></A
><P
> Rather than look at <TT
CLASS="STRUCTNAME"
>pg_statistic</TT
> directly,
it's better to look at its view
<A
HREF="view-pg-stats.html"
><TT
CLASS="STRUCTNAME"
>pg_stats</TT
></A
>
when examining the statistics manually. <TT
CLASS="STRUCTNAME"
>pg_stats</TT
>
is designed to be more easily readable. Furthermore,
<TT
CLASS="STRUCTNAME"
>pg_stats</TT
> is readable by all, whereas
<TT
CLASS="STRUCTNAME"
>pg_statistic</TT
> is only readable by a superuser.
(This prevents unprivileged users from learning something about
the contents of other people's tables from the statistics. The
<TT
CLASS="STRUCTNAME"
>pg_stats</TT
> view is restricted to show only
rows about tables that the current user can read.)
For example, we might do:
</P><PRE
CLASS="SCREEN"
>SELECT attname, n_distinct, most_common_vals FROM pg_stats WHERE tablename = 'road';
attname | n_distinct | most_common_vals
---------+------------+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
name | -0.467008 | {"I- 580 Ramp","I- 880 Ramp","Sp Railroad ","I- 580 ","I- 680 Ramp","I- 80 Ramp","14th St ","5th St ","Mission Blvd","I- 880 "}
thepath | 20 | {"[(-122.089,37.71),(-122.0886,37.711)]"}
(2 rows)</PRE
><P>
</P
><P
> The amount of information stored in <TT
CLASS="STRUCTNAME"
>pg_statistic</TT
>
by <TT
CLASS="COMMAND"
>ANALYZE</TT
>, in particular the maximum number of entries in the
<TT
CLASS="STRUCTFIELD"
>most_common_vals</TT
> and <TT
CLASS="STRUCTFIELD"
>histogram_bounds</TT
>
arrays for each column, can be set on a
column-by-column basis using the <TT
CLASS="COMMAND"
>ALTER TABLE SET STATISTICS</TT
>
command, or globally by setting the
<A
HREF="runtime-config-query.html#GUC-DEFAULT-STATISTICS-TARGET"
>default_statistics_target</A
> configuration variable.
The default limit is presently 10 entries. Raising the limit
might allow more accurate planner estimates to be made, particularly for
columns with irregular data distributions, at the price of consuming
more space in <TT
CLASS="STRUCTNAME"
>pg_statistic</TT
> and slightly more
time to compute the estimates. Conversely, a lower limit might be
sufficient for columns with simple data distributions.
</P
><P
> Further details about the planner's use of statistics can be found in
<A
HREF="planner-stats-details.html"
>Chapter 55</A
>.
</P
></DIV
><DIV
CLASS="NAVFOOTER"
><HR
ALIGN="LEFT"
WIDTH="100%"><TABLE
SUMMARY="Footer navigation table"
WIDTH="100%"
BORDER="0"
CELLPADDING="0"
CELLSPACING="0"
><TR
><TD
WIDTH="33%"
ALIGN="left"
VALIGN="top"
><A
HREF="using-explain.html"
ACCESSKEY="P"
>Prev</A
></TD
><TD
WIDTH="34%"
ALIGN="center"
VALIGN="top"
><A
HREF="index.html"
ACCESSKEY="H"
>Home</A
></TD
><TD
WIDTH="33%"
ALIGN="right"
VALIGN="top"
><A
HREF="explicit-joins.html"
ACCESSKEY="N"
>Next</A
></TD
></TR
><TR
><TD
WIDTH="33%"
ALIGN="left"
VALIGN="top"
>Using <TT
CLASS="COMMAND"
>EXPLAIN</TT
></TD
><TD
WIDTH="34%"
ALIGN="center"
VALIGN="top"
><A
HREF="performance-tips.html"
ACCESSKEY="U"
>Up</A
></TD
><TD
WIDTH="33%"
ALIGN="right"
VALIGN="top"
>Controlling the Planner with Explicit <TT
CLASS="LITERAL"
>JOIN</TT
> Clauses</TD
></TR
></TABLE
></DIV
></BODY
></HTML
>
distrib
>
Mandriva
>
2008.1
>
x86_64
>
media
>
main-testing
>
by-pkgid
>
bab02a23fa9f3df8d66a9a3231b50245
>
files
>
402
