summaryrefslogtreecommitdiff
path: root/doc/src
diff options
context:
space:
mode:
Diffstat (limited to 'doc/src')
-rw-r--r--doc/src/sgml/installation.sgml16
-rw-r--r--doc/src/sgml/storage.sgml8
2 files changed, 21 insertions, 3 deletions
diff --git a/doc/src/sgml/installation.sgml b/doc/src/sgml/installation.sgml
index a9990023466..95a3f10be67 100644
--- a/doc/src/sgml/installation.sgml
+++ b/doc/src/sgml/installation.sgml
@@ -1,4 +1,4 @@
-<!-- $PostgreSQL: pgsql/doc/src/sgml/installation.sgml,v 1.303 2008/03/06 21:37:33 momjian Exp $ -->
+<!-- $PostgreSQL: pgsql/doc/src/sgml/installation.sgml,v 1.304 2008/03/10 20:06:27 tgl Exp $ -->
<chapter id="installation">
<title><![%standalone-include[<productname>PostgreSQL</>]]>
@@ -1026,6 +1026,20 @@ su - postgres
</varlistentry>
<varlistentry>
+ <term><option>--disable-segmented-files</option></term>
+ <listitem>
+ <para>
+ Store large tables as single operating-system files, rather than
+ dividing them into 1GB segments as is the default. This option
+ is ignored unless the operating system has <quote>largefile</>
+ support (which most do, nowadays). It can be helpful to reduce
+ the number of file descriptors consumed when working with very
+ large tables.
+ </para>
+ </listitem>
+ </varlistentry>
+
+ <varlistentry>
<term><option>--disable-spinlocks</option></term>
<listitem>
<para>
diff --git a/doc/src/sgml/storage.sgml b/doc/src/sgml/storage.sgml
index fe9ae611bf1..7ba0c1e343f 100644
--- a/doc/src/sgml/storage.sgml
+++ b/doc/src/sgml/storage.sgml
@@ -1,4 +1,4 @@
-<!-- $PostgreSQL: pgsql/doc/src/sgml/storage.sgml,v 1.21 2007/11/23 00:24:12 ishii Exp $ -->
+<!-- $PostgreSQL: pgsql/doc/src/sgml/storage.sgml,v 1.22 2008/03/10 20:06:27 tgl Exp $ -->
<chapter id="storage">
@@ -138,10 +138,14 @@ Avoid assuming that filenode and table OID are the same.
</caution>
<para>
-When a table or index exceeds 1 GB, it is divided into gigabyte-sized
+When a table or index exceeds 1 GB, it is normally divided into gigabyte-sized
<firstterm>segments</>. The first segment's file name is the same as the
filenode; subsequent segments are named filenode.1, filenode.2, etc.
This arrangement avoids problems on platforms that have file size limitations.
+(But if the platform does not have such a limitation, and
+<option>--disable-segmented-files</option> was specified when
+<productname>PostgreSQL</> was built, then each table or index is stored
+as a single file, without segmentation.)
The contents of tables and indexes are discussed further in
<xref linkend="storage-page-layout">.
</para>