[gmx-users] Gromacs on SGI Altix
mfyang at gmail.com
Tue Jan 3 17:21:35 CET 2006
Florian Haberl wrote:
> On Tuesday 03 January 2006 16:08, Mingfeng Yang wrote:
>> Happy new year to each gromacs user!
>> I just finished installation of Gromacs 3.3 on our Itanium2 based SGI
>> Altix. My configuration option is
>> |./configure --prefix=/home/myang/usr/local/gromacs_intel --enable-mpi
>> CC=icc F77=ifort|
>> The intel C++ compiler 9.021 and intel fortran compiler 9.021 were used.
>> The OS is Red Hat Enterprise WS 3.0.
>> The parallel efficiency is kind of disappointing. For the benchmark case
>> of villin, it took 97 seconds with one CPU, 55 seconds with two CPUs,
>> and 35 seconds for 4 CPUs. From the benchmark pages of Gromacs website,
>> I saw the parallel efficiency can go beyond 100% under 8 CPUs. Is there
>> anything I missed? What supposed to be the best options when install
>> gromacs on Itanium2 based machine?
> I think that d.villin is not the right benchmark for more cpus and measering
> parallel speedup.
> I would prefer here d.dppc as shown on gromacs webpage. Reason for this bad
> scalling is is small system size.
Thank you, Florian! I will try d.dppc. I have another question. Is
Gromacs specifically optimized for x86 machines, and the power of
Itanium2 is not fully used. Because for Amber8, my Itanium2 is almost
two times faster than my opteron, but for Gromacs, opteron is slightly
(~1.2 times) faster than Itanium2. Other factors are almost same. Just
due to the reason that there is no assembly loops optimization for IA64?
>> Thank you for any input!
>> gmx-users mailing list
>> gmx-users at gromacs.org
>> Please don't post (un)subscribe requests to the list. Use the
>> www interface or send it to gmx-users-request at gromacs.org.
More information about the gromacs.org_gmx-users