Dear Carsten,
Thanks for the advice!
The thing is that considering the price differences, we could buy 2-3
AMD machines (48 core) for each 1 Intel machine (32) core.
I'm only curious about how much faster Intel will be. Side to side,
amd 48core vs intel 32core, would make us talk about 2x differences in
ns/day?
Thanks again.
Regards,
David
Message: 1
Date: Thu, 20 Jan 2011 13:45:54 +0100
From: Carsten Kutzner <ckut...@gwdg.de>
Subject: Re: [gmx-users] 4 x Opteron 12-core or 4 x Xeon 8-core ?
To: "Discussion list for GROMACS users" <gmx-users@gromacs.org>
Message-ID: <dba5ecbb-d1f3-4559-a849-fb8cf16c9...@gwdg.de>
Content-Type: text/plain; charset=us-ascii
Hi David,
On Jan 20, 2011, at 1:21 PM, David McGiven wrote:
Dear Gromacs Users,
We're going to buy a new server for HPC. It is going to run mainly
Gromacs calculations.
Regarding Gromacs performance, I'm wondering which one, you Gromacs
users and developers, think will be faster.
AMD Server : 4 x AMD Opteron 6176 12-core 2.3 Ghz + 96GB Memory
(2GB / core)
Intel Server : 4 x Intel Xeon 8-core 2.66 Ghz + 64 GB RAM (2GB /
core)
We normally run ~100k atom systems with PME and explicit water.
Which one would you recommend ?
Also, of course, AMD Server is cheaper. But we are mainly
interested on performance.
If you have a fixed amount of money, you will get the most ns/day
if you buy the AMD Magny Cours machines. Each one will be slower
compared to the
Intel server but you will get more servers altogether, thus more total
performance. If you can only buy a single server and you do not care
about
what it costs, the Intel will be faster for shure.
Note that you do not need 2 GB/core for 100k atoms MD systems if you
run
Gromacs. Half of it will be more than enough.
Carsten
Thanks.
Best Regards,
David
--
gmx-users mailing list gmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/Search
before posting!
Please don't post (un)subscribe requests to the list. Use thewww
interface or send it to gmx-users-requ...@gromacs.org.
Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
--
Dr. Carsten Kutzner
Max Planck Institute for Biophysical Chemistry
Theoretical and Computational Biophysics
Am Fassberg 11, 37077 Goettingen, Germany
Tel. +49-551-2012313, Fax: +49-551-2012302
http://www.mpibpc.mpg.de/home/grubmueller/ihp/ckutzne
--
gmx-users mailing list gmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the
www interface or send it to gmx-users-requ...@gromacs.org.
Can't post? Read http://www.gromacs.org/Support/Mailing_Lists