From: | mlw <markw(at)mohawksoft(dot)com> |
---|---|
To: | PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org> |
Subject: | 8K vs 16K block size report |
Date: | 2002-04-26 02:16:00 |
Message-ID: | 3CC8B860.432D1A9C@mohawksoft.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
I have been doing some benchmarking. The idea is to measure the difference
between PostgreSQL 7.2.1 compiled for an 8K block or a 16K block. pgbench is
almost useless for measuring changes in performance, I can't seem to get any
real consistency, so I used the osdb to measure performance.
In the 16K block configuration, I reduced the number of blocks by half, so as
to keep caching memory similarly sized.
The machine:
RedHat 7.2, dual PIII 650, 1G ram, 2 IBM scsi 18G drives.
uname -a:
Linux slave1.mohawksoft.com 2.4.7-10smp #1 SMP Thu Sep 6 17:09:31 EDT 2001 i686
unknown
mount:
[markw(at)slave1 markw]$ mount
/dev/sda3 on / type ext3 (rw)
none on /proc type proc (rw)
usbdevfs on /proc/bus/usb type usbdevfs (rw)
/dev/sda1 on /boot type ext3 (rw)
/dev/sda4 on /u01 type ext2 (rw)
/dev/sdb1 on /u02 type ext2 (rw)
none on /dev/pts type devpts (rw,gid=5,mode=620)
none on /dev/shm type tmpfs (rw)
The 8K data directory:
[postgres(at)slave1 data]$ pwd
/home/postgres/data
[postgres(at)slave1 data]$ ls -l
total 40
lrwxrwxrwx 1 root root 9 Mar 6 07:21 base -> /u02/base
drwx------ 2 postgres postgres 4096 Apr 25 21:51 global
drwx------ 2 postgres postgres 4096 Apr 25 21:29 pg_clog
-rw------- 1 postgres postgres 10068 Mar 6 07:26 pg_hba.conf
-rw------- 1 postgres postgres 1250 Mar 6 07:20 pg_ident.conf
-rw------- 1 postgres postgres 4 Mar 6 07:20 PG_VERSION
lrwxrwxrwx 1 root root 12 Mar 6 07:22 pg_xlog -> /u01/pg_xlog
-rw------- 1 postgres postgres 3095 Apr 24 19:23 postgresql.conf
-rw------- 1 postgres postgres 56 Apr 25 20:54 postmaster.opts
-rw------- 1 postgres postgres 44 Apr 25 20:54 postmaster.pid
The 16K data directory:
[postgres(at)slave1 data16kb]$ pwd
/home/postgres/data16kb
[postgres(at)slave1 data16kb]$ ls -l
total 36
lrwxrwxrwx 1 postgres postgres 13 Apr 25 08:47 base -> /u02/base16kb
drwx------ 2 postgres postgres 4096 Apr 25 20:53 global
drwx------ 2 postgres postgres 4096 Apr 25 20:33 pg_clog
-rw------- 1 postgres postgres 10068 Apr 25 08:44 pg_hba.conf
-rw------- 1 postgres postgres 1250 Apr 25 08:43 pg_ident.conf
-rw------- 1 postgres postgres 4 Apr 25 08:43 PG_VERSION
lrwxrwxrwx 1 postgres postgres 16 Apr 25 08:46 pg_xlog ->
/u01/pg_xlog16kb
-rw------- 1 postgres postgres 3095 Apr 25 08:44 postgresql.conf
-rw------- 1 postgres postgres 64 Apr 25 14:59 postmaster.opts
Attachment | Content-Type | Size |
---|---|---|
osdb16.log | text/plain | 6.0 KB |
osdb8.log | text/plain | 6.0 KB |
pgconf16k.conf | text/plain | 3.0 KB |
pgconf8k.conf | text/plain | 3.0 KB |
From | Date | Subject | |
---|---|---|---|
Next Message | Bruce Momjian | 2002-04-26 02:18:47 | Re: Sequential Scan Read-Ahead |
Previous Message | Marc G. Fournier | 2002-04-26 01:56:57 | Re: Vote totals for SET in aborted transaction |