mirror of
https://git.postgresql.org/git/postgresql.git
synced 2024-09-04 12:47:23 +02:00
Improve documentation about use of Linux huge pages.
Show how to get the system's huge page size, rather than misleadingly referring to PAGE_SIZE (which is usually understood to be the regular page size). Show how to confirm whether huge pages have been allocated. Minor wordsmithing. Back-patch to 9.4 where this section appeared.
This commit is contained in:
parent
8acc15eaa1
commit
13d5dd5095
@ -1375,53 +1375,67 @@ export PG_OOM_ADJUST_VALUE=0
|
|||||||
</sect2>
|
</sect2>
|
||||||
|
|
||||||
<sect2 id="linux-huge-pages">
|
<sect2 id="linux-huge-pages">
|
||||||
<title>Linux huge pages</title>
|
<title>Linux Huge Pages</title>
|
||||||
|
|
||||||
<para>
|
<para>
|
||||||
Using huge pages reduces overhead when using large contiguous chunks of
|
Using huge pages reduces overhead when using large contiguous chunks of
|
||||||
memory, like <productname>PostgreSQL</productname> does. To enable this
|
memory, as <productname>PostgreSQL</productname> does, particularly when
|
||||||
|
using large values of <xref linkend="guc-shared-buffers">. To use this
|
||||||
feature in <productname>PostgreSQL</productname> you need a kernel
|
feature in <productname>PostgreSQL</productname> you need a kernel
|
||||||
with <varname>CONFIG_HUGETLBFS=y</varname> and
|
with <varname>CONFIG_HUGETLBFS=y</varname> and
|
||||||
<varname>CONFIG_HUGETLB_PAGE=y</varname>. You also have to tune the system
|
<varname>CONFIG_HUGETLB_PAGE=y</varname>. You will also have to adjust
|
||||||
setting <varname>vm.nr_hugepages</varname>. To estimate the number of
|
the kernel setting <varname>vm.nr_hugepages</varname>. To estimate the
|
||||||
necessary huge pages start <productname>PostgreSQL</productname> without
|
number of huge pages needed, start <productname>PostgreSQL</productname>
|
||||||
huge pages enabled and check the <varname>VmPeak</varname> value from the
|
without huge pages enabled and check the
|
||||||
proc file system:
|
postmaster's <varname>VmPeak</varname> value, as well as the system's
|
||||||
|
huge page size, using the <filename>/proc</> file system. This might
|
||||||
|
look like:
|
||||||
<programlisting>
|
<programlisting>
|
||||||
$ <userinput>head -1 /path/to/data/directory/postmaster.pid</userinput>
|
$ <userinput>head -1 $PGDATA/postmaster.pid</userinput>
|
||||||
4170
|
4170
|
||||||
$ <userinput>grep ^VmPeak /proc/4170/status</userinput>
|
$ <userinput>grep ^VmPeak /proc/4170/status</userinput>
|
||||||
VmPeak: 6490428 kB
|
VmPeak: 6490428 kB
|
||||||
|
$ <userinput>grep ^Hugepagesize /proc/meminfo</userinput>
|
||||||
|
Hugepagesize: 2048 kB
|
||||||
</programlisting>
|
</programlisting>
|
||||||
<literal>6490428</literal> / <literal>2048</literal>
|
<literal>6490428</literal> / <literal>2048</literal> gives approximately
|
||||||
(<varname>PAGE_SIZE</varname> is <literal>2MB</literal> in this case) are
|
<literal>3169.154</literal>, so in this example we need at
|
||||||
roughly <literal>3169.154</literal> huge pages, so you will need at
|
least <literal>3170</literal> huge pages, which we can set with:
|
||||||
least <literal>3170</literal> huge pages:
|
|
||||||
<programlisting>
|
<programlisting>
|
||||||
$ <userinput>sysctl -w vm.nr_hugepages=3170</userinput>
|
$ <userinput>sysctl -w vm.nr_hugepages=3170</userinput>
|
||||||
</programlisting>
|
</programlisting>
|
||||||
Sometimes the kernel is not able to allocate the desired number of huge
|
A larger setting would be appropriate if other programs on the machine
|
||||||
pages, so it might be necessary to repeat that command or to reboot. Don't
|
also need huge pages. Don't forget to add this setting
|
||||||
forget to add an entry to <filename>/etc/sysctl.conf</filename> to persist
|
to <filename>/etc/sysctl.conf</filename> so that it will be reapplied
|
||||||
this setting through reboots.
|
after reboots.
|
||||||
</para>
|
</para>
|
||||||
|
|
||||||
<para>
|
<para>
|
||||||
It is also necessary to give the database server operating system
|
Sometimes the kernel is not able to allocate the desired number of huge
|
||||||
|
pages immediately, so it might be necessary to repeat the command or to
|
||||||
|
reboot. (Immediately after a reboot, most of the machine's memory
|
||||||
|
should be available to convert into huge pages.) To verify the huge
|
||||||
|
page allocation situation, use:
|
||||||
|
<programlisting>
|
||||||
|
$ <userinput>grep Huge /proc/meminfo</userinput>
|
||||||
|
</programlisting>
|
||||||
|
</para>
|
||||||
|
|
||||||
|
<para>
|
||||||
|
It may also be necessary to give the database server's operating system
|
||||||
user permission to use huge pages by setting
|
user permission to use huge pages by setting
|
||||||
<varname>vm.hugetlb_shm_group</> via <application>sysctl</>, and
|
<varname>vm.hugetlb_shm_group</> via <application>sysctl</>, and/or
|
||||||
permission to lock memory with <command>ulimit -l</>.
|
give permission to lock memory with <command>ulimit -l</>.
|
||||||
</para>
|
</para>
|
||||||
|
|
||||||
<para>
|
<para>
|
||||||
The default behavior for huge pages in
|
The default behavior for huge pages in
|
||||||
<productname>PostgreSQL</productname> is to use them when possible and
|
<productname>PostgreSQL</productname> is to use them when possible and
|
||||||
to fallback to normal pages when failing. To enforce the use of huge
|
to fall back to normal pages when failing. To enforce the use of huge
|
||||||
pages, you can set
|
pages, you can set <xref linkend="guc-huge-pages">
|
||||||
<link linkend="guc-huge-pages"><varname>huge_pages</varname></link>
|
to <literal>on</literal> in <filename>postgresql.conf</>.
|
||||||
to <literal>on</literal>. Note that in this case
|
Note that with this setting <productname>PostgreSQL</> will fail to
|
||||||
<productname>PostgreSQL</productname> will fail to start if not enough huge
|
start if not enough huge pages are available.
|
||||||
pages are available.
|
|
||||||
</para>
|
</para>
|
||||||
|
|
||||||
<para>
|
<para>
|
||||||
|
Loading…
Reference in New Issue
Block a user