Re: Extreme high load averages - Mailing list pgsql-performance

From Shridhar Daithankar
Subject Re: Extreme high load averages
Date
Msg-id 200307061604.48919.shridhar_daithankar@nospam.persistent.co.in
Whole thread Raw
In response to Re: Extreme high load averages  (Martin Foster <martin@ethereal-realms.org>)
Responses Re: Extreme high load averages
List pgsql-performance
On Sunday 06 July 2003 15:56, Martin Foster wrote:
> effective_cache_size seems interesting, though the description is
> somewhat lacking.  Is this related to the swap partition and how much of
> it will be used by PostgreSQL?   If I am correct, this should be fairly
> low? Martin Foster

It gives hint to psotgresql how much file system cache is available in the
system.

You have 1GB memory and your application requirement does not exceed 400MB. So
OS can use roughly 600MB for file system cache. In that case you can set this
parameter to 400MB cache to leave room for other application in FS cache.

IIRC, BSD needs sysctl tuning to make more memory available for FS cache other
wise they max out at 300MB.

Roughly this setting should be (total memory -application
requirement)*(0.7/0.8)

I guess that high kernel load you are seeing due to increased interaction
between postgresql and OS when data is swapped to/fro in shared memory. If OS
cache does well, postgresql should reduce this interaction as well.


BTW, since you have IDE disks, heavy disk activity can eat CPU as well. Is
your disk bandwidth totally maxed out? Check with vmstat or whatever
equivalent you have on BSD.

 Shridhar


pgsql-performance by date:

Previous
From: Martin Foster
Date:
Subject: Re: Extreme high load averages
Next
From: Shridhar Daithankar
Date:
Subject: Another POC initdb patch