|
|
|
@ -9,9 +9,10 @@ |
|
|
|
|
|
|
|
|
|
<para> |
|
|
|
|
<application>pg_test_fsync</> is intended to give you a reasonable |
|
|
|
|
idea of what the fastest fsync_method is on your specific system, |
|
|
|
|
idea of what the fastest <xref linkend="guc-wal-sync-method"> is on your |
|
|
|
|
specific system, |
|
|
|
|
as well as supplying diagnostic information in the event of an |
|
|
|
|
identified I/O problem. However, differences shown by test_fsync |
|
|
|
|
identified I/O problem. However, differences shown by pg_test_fsync |
|
|
|
|
might not make any difference in real database throughput, especially |
|
|
|
|
since many database servers are not speed-limited by their transaction |
|
|
|
|
logs. |
|
|
|
@ -20,11 +21,13 @@ |
|
|
|
|
<sect2> |
|
|
|
|
<title>Usage</title> |
|
|
|
|
|
|
|
|
|
<sect2> |
|
|
|
|
<title><application>pg_test_fsync</> Options</title> |
|
|
|
|
<synopsis> |
|
|
|
|
pg_test_fsync [options] |
|
|
|
|
</synopsis> |
|
|
|
|
|
|
|
|
|
<para> |
|
|
|
|
<application>pg_test_fsync</application> accepts the following command-line arguments: |
|
|
|
|
<application>pg_test_fsync</application> accepts the following |
|
|
|
|
command-line options: |
|
|
|
|
|
|
|
|
|
<variablelist> |
|
|
|
|
|
|
|
|
@ -33,9 +36,10 @@ |
|
|
|
|
<term><option>--filename</option></term> |
|
|
|
|
<listitem> |
|
|
|
|
<para> |
|
|
|
|
Specifies the filename for test. The file should be in the |
|
|
|
|
same file system as the <filename>pg_xlog</> directory. |
|
|
|
|
<filename>/pg_xlog</> contains the <acronym>WAL</> files.) |
|
|
|
|
Specifies the filename to write test data in. |
|
|
|
|
This file should be in the same file system that the |
|
|
|
|
<filename>pg_xlog</> directory is or will be placed in. |
|
|
|
|
(<filename>pg_xlog</> contains the <acronym>WAL</> files.) |
|
|
|
|
The default is <filename>test_fsync.out</> in the current |
|
|
|
|
directory. |
|
|
|
|
</para> |
|
|
|
@ -48,8 +52,8 @@ |
|
|
|
|
<listitem> |
|
|
|
|
<para> |
|
|
|
|
Specifies the number of operations per test. The more operations |
|
|
|
|
per test, the greater the test's accuracy. The default is |
|
|
|
|
2000. |
|
|
|
|
per test, the greater the test's accuracy, but the longer it takes |
|
|
|
|
to run. The default is 2000. |
|
|
|
|
</para> |
|
|
|
|
</listitem> |
|
|
|
|
</varlistentry> |
|
|
|
|