Re: disk performance benchmarks - Mailing list pgsql-general

From Michael Paesold
Subject Re: disk performance benchmarks
Date
Msg-id 003801c49b07$f33b6990$ad01a8c0@zaphod
Whole thread Raw
In response to disk performance benchmarks  ("Shane Wright" <shiversraa@gmail.com>)
Responses Re: disk performance benchmarks  ("Jeffrey W. Baker" <jwbaker@acm.org>)
List pgsql-general
Jeffrey W. Baker wrote:

> Current issue:
>
> A dual 64-bit Opteron 244 machine with 8GB main memory, two 4-disk RAID5
> arrays (one for database, one for xlogs).  PG's config is extremely
> generous, and in isolated benchmarks it's very fast.

It depends on the controller, but usually I would expect a better
performance if xlogs are just on a two-disk mirror and the rest of the disks
for data (6 splindles instead of 4 then).

I don't think RAID5 is a benefit for xlogs.

Regards,
Michael Paesold

> But, in reality, performance is abyssmal.  There's something about what
> PG does inside commits and checkpoints that sends Linux into a catatonic
> state.  For instance here's a snapshot of vmstat during a parallel heavy
> select/insert load:
...


pgsql-general by date:

Previous
From: Marcel Boscher
Date:
Subject: Tsearch2 adding additional dictionaries
Next
From: "Katsaros Kwn/nos"
Date:
Subject: Problems with SPI memory management