From: bh yuan <bhyuan@gmail.com>
To: pgsql-performance@postgresql.org
Subject: Question about shared_buffers and cpu usage
Date: Thu, 21 Feb 2008 14:13:50 +0900
Message-ID: <3c7b84c00802202113x3a705e34u9d9cf307fdde8048@mail.gmail.com> (raw)
Hi
I am using Postgres8.3 on 8G memory , Xeon X5355 Quad Core x 2
processer RH5 machine with 10G data. (with some table which have
about 2,000,000~ 5,000,000 rows )
I have two quesion.
1. how to set the shared_buffers and other postgresql.conf parameter
for best performance?
I only run the Postgres8.3 on the machine so I set the shared_buffers
= 7168MB (7G)
But somebody said it is too big, so confused.
The memory info is that
-----------------------------------------------------
MemTotal: 8177484 kB
MemFree: 313336 kB
Buffers: 112700 kB
Cached: 7008160 kB
SwapCached: 210832 kB
Active: 7303660 kB
Inactive: 402088 kB
HighTotal: 0 kB
HighFree: 0 kB
LowTotal: 8177484 kB
LowFree: 313336 kB
SwapTotal: 8385920 kB
SwapFree: 7415768 kB
Dirty: 908 kB
Writeback: 0 kB
AnonPages: 28312 kB
Mapped: 2163912 kB
Slab: 99396 kB
PageTables: 13004 kB
NFS_Unstable: 0 kB
Bounce: 0 kB
CommitLimit: 12474660 kB
Committed_AS: 8169440 kB
VmallocTotal: 34359738367 kB
VmallocUsed: 267136 kB
VmallocChunk: 34359470587 kB
HugePages_Total: 0
HugePages_Free: 0
HugePages_Rsvd: 0
Hugepagesize: 2048 kB
-----------------------------------------------
2 I have 8 core cpu ,but It seems that one sql can only use 1 core.
Can I use more core to execute one sql to optimize the speed ?
Thanks
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Reply to all the recipients using the --to and --cc options:
reply via email
To: pgsql-performance@postgresql.org
Cc: bhyuan@gmail.com
Subject: Re: Question about shared_buffers and cpu usage
In-Reply-To: <3c7b84c00802202113x3a705e34u9d9cf307fdde8048@mail.gmail.com>
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox