[gmx-users] High speed performance

[Kutzner, Carsten]

> On 04 Aug 2015, at 00:02, Asmaa El khodary <asma.moh_990 at aucegypt.edu> wrote:
> 
> Yes, I’m running the simulation on a cluster using 5 computational nodes,
In that case I would run on a single node with 8 cores first so that you
can compare the parallel performance to a single node and can derive how
high the parallel efficiency is. That will depend a lot on your network.
This is worth the effort, as sometimes using less nodes results in higher
performance :)

Carsten


> each one has 8 cores, so the total number of cores is 40. The module I'm
> using is GROMACS/4.6.5-goolf-1.4.10-hybrid.
> 
> On Tue, Aug 4, 2015 at 12:45 AM, Szilárd Páll <pall.szilard at gmail.com>
> wrote:
> 
>> 40 cores? Are you sure it's not two 10-core CPUs with HyperThreading = 2x20
>> threads supported?
>> 
>> --
>> Szilárd
>> 
>> On Mon, Aug 3, 2015 at 3:19 PM, Asmaa El khodary <
>> asma.moh_990 at aucegypt.edu>
>> wrote:
>> 
>>> Dr Carsten
>>> Thank you so much for your reply. Yes, you are right, it's a 40 core CPU.
>>> I'm using 2 fs time step with Particle Mesh Ewald for long-range
>>> electrostatics. I've attached the parameter file I'm using for the
>>> production run.
>>> Best Regards
>>> Asma
>>> 
>>> On Mon, Aug 3, 2015 at 4:03 PM, Asmaa El khodary <
>>> asma.moh_990 at aucegypt.edu>
>>> wrote:
>>> 
>>>> Hi,
>>>> I'm currently simulating a protein consists of 2776 atoms with 173
>>>> residues using CPU with 5 nodes, 40 core processors. The performance of
>>> the
>>>> simulation is 70 ns per day (50 ns simulation takes almost 18 hours).
>> Is
>>>> this realistic or not?
>>>> Thanks in Advance
>>>> Asma
>>>> 
>>> 
>>> --
>>> Gromacs Users mailing list
>>> 
>>> * Please search the archive at
>>> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
>>> posting!
>>> 
>>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>> 
>>> * For (un)subscribe requests visit
>>> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
>>> send a mail to gmx-users-request at gromacs.org.
>>> 
>> --
>> Gromacs Users mailing list
>> 
>> * Please search the archive at
>> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
>> posting!
>> 
>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>> 
>> * For (un)subscribe requests visit
>> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
>> send a mail to gmx-users-request at gromacs.org.
> -- 
> Gromacs Users mailing list
> 
> * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!
> 
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> 
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-request at gromacs.org.

--
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/grubmueller/kutzner
http://www.mpibpc.mpg.de/grubmueller/sppexa