Re: Poor performance on seq scan - Mailing list pgsql-performance

From Ivan Voras
Subject Re: Poor performance on seq scan
Date
Msg-id 450871C9.1010709@fer.hr
Whole thread Raw
In response to Poor performance on seq scan  (Laszlo Nagy <gandalf@designaproduct.biz>)
List pgsql-performance
pgsql-performance-owner@postgresql.org wrote:

> This board has Intel chipset. I cannot remember the exact type but it
> was not in the low end category.
> dmesg says:
>
> <Intel ICH7 SATA300 controller>
> kernel: ad4: 152626MB <SAMSUNG HD160JJ ZM100-33> at ata2-master SATA150
> kernel: ad4: 152627MB <SAMSUNG HD160JJ ZM100-33> at ata3-master SATA150

There have been reported problems with ICH7 on FreeBSD mailing lists,
though I can't find any that affect performance.

> Components: 2
> Balance: round-robin
> Slice: 4096

See if changing balance algorithm to "split", and slice size to 8192 or
more, while keeping vfs.read_max to 16 or more helps your performance.

(e.g. gmirror configure -b split -s 8192 gm0)

Also, how is your file system mounted? (what does output from 'mount' say?)

pgsql-performance by date:

Previous
From: "Joshua Marsh"
Date:
Subject: Re: Performance With Joins on Large Tables
Next
From: Tom Lane
Date:
Subject: Re: Performance With Joins on Large Tables