From: | Greg Smith <greg(at)2ndquadrant(dot)com> |
---|---|
To: | "pgsql-performance(at)postgresql(dot)org" <pgsql-performance(at)postgresql(dot)org> |
Subject: | Re: Memory speed testing |
Date: | 2010-09-27 16:53:26 |
Message-ID: | 4CA0CC06.4040809@2ndquadrant.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-performance |
Last bit of promotion here for this project:
http://github.com/gregs1104/stream-scaling has been updated with a few
more results anonymously submitted by list members here. The updated
chart really fleshes out how the RAM scaling works on the older
multi-socket AMD servers now. That's something that's been described on
this list for years, but never really measured this clearly before to my
knowledge. I have 4 and 8 socket examples in there now, showing
DDR2/667 and DDR2/800.
Also have some initial numbers on the DDR-3 based Phenom II processors
too, which fit right in the middle of the Intel samples. They look like
another round of good bang for the buck midrange desktop parts typical
for AMD.
--
Greg Smith, 2ndQuadrant US greg(at)2ndQuadrant(dot)com Baltimore, MD
PostgreSQL Training, Services and Support www.2ndQuadrant.us
Author, "PostgreSQL 9.0 High Performance" Pre-ordering at:
https://www.packtpub.com/postgresql-9-0-high-performance/book
From | Date | Subject | |
---|---|---|---|
Next Message | Robert Haas | 2010-09-27 17:30:03 | Re: how to enforce index sub-select over filter+seqscan |
Previous Message | adrian.kitchingman | 2010-09-27 00:06:34 | Re: postgresql-9.0 Windows service stops after database transaction |