Re: Xeon twice the performance of opteron - Mailing list pgsql-performance

From Claudio Freire
Subject Re: Xeon twice the performance of opteron
Date
Msg-id AANLkTikSxo6s8D65-MZkv_D7fwbLe2=9Pa13HGAS8ORV@mail.gmail.com
Whole thread Raw
In response to Xeon twice the performance of opteron  (Jeff <threshar@torgo.978.org>)
List pgsql-performance
On Thu, Mar 17, 2011 at 1:42 PM, J Sisson <sisson.j@gmail.com> wrote:
> On Thu, Mar 17, 2011 at 10:13 AM, Jeff <threshar@torgo.978.org> wrote:
>> hey folks,
>>
>> Running into some odd performance issues between a few of our db boxes.
>
> We've noticed similar results both in OLTP and data warehousing conditions here.
>
> Opteron machines just seem to lag behind *especially* in data
> warehousing.  Smaller
> cache for sorting/etc... is what I'd always chalked it up to, but I'm
> open to other theories
> if they exist.

It's my theory as well - you know, this could be solved by JITting
complex expressions.

Bad cache behavior in application often comes as a side-effect of
interpreted execution (in this case, of expressions, conditions,
functions). A JIT usually solves this cache inefficiency.

I know, adding any kind of JIT to pg could be a major task.

pgsql-performance by date:

Previous
From: J Sisson
Date:
Subject: Re: Xeon twice the performance of opteron
Next
From: Craig James
Date:
Subject: Re: Xeon twice the performance of opteron