Re: Warm-cache prefetching - Mailing list pgsql-hackers

From Qingqing Zhou
Subject Re: Warm-cache prefetching
Date
Msg-id Pine.LNX.4.58.0512091402440.22607@eon.cs
Whole thread Raw
In response to Re: Warm-cache prefetching  ("Min Xu (Hsu)" <xu@cs.wisc.edu>)
List pgsql-hackers
On Thu, 8 Dec 2005, Min Xu (Hsu) wrote:

> Perhaps because P4 is already doing H/W prefetching?
>
> http://www.tomshardware.com/2000/11/20/intel/page5.html
>
> I ran the test program on an opteron 2.2G:
> Got slowdown.
>
> I ran it on a PIII 650M:
> looks like some speedup to me.
>

Ok, I see ... so this optimization relies on the CPU type and even speed.
If many programs are fighting for L2 cache, then things will be screwed
up.

Regards,
Qingqing


pgsql-hackers by date:

Previous
From: Alvaro Herrera
Date:
Subject: Re: Min Xid problem proposal
Next
From: Tom Lane
Date:
Subject: Re: Min Xid problem proposal