[gmx-users] Is version 5.0 generating portable binaries?

[jkrieger at mrc-lmb.cam.ac.uk]

Can you not find out what acceleration types you cluster has from the cluster admin and select nodes appropriately? I do this on our cluster by providing qsub with the flag -l node_type=m620

On 31 Jan 2014, at 01:40, Mark Abraham <mark.j.abraham at gmail.com> wrote:

> On Jan 30, 2014 11:57 PM, "Raffaella D'Auria" <rdauria at ucla.edu> wrote:
>> 
>> Hello All,
>> 
>> I have just performed a fresh installation of gromacs version 4.6.5 on a
> heterogeneous linux cluster (running centos 6.4) . After the installation I
> see that the binaries, for say mdrun,  work on all the machines with
> hardware similar to the one on which they were installed (namely AMD
> processor such as: Quad-Core AMD Opteron Processor 2354) but produce an
> "illegal instruction (core dumped)" when run on nodes with intel processors
> (such as: Intel Xeon CPU E5335  @ 2.00GHz).
>> 
>> As suggested in the documentation (
> http://www.gromacs.org/Documentation/Acceleration_and_parallelization#CPU_acceleration.3a_SSE.2c_AVX.2c_etc)
> I have tried to compile with GMX_CPU_ACCELERATION set to "none" (I have
> also turned down the optimization levels) with always the same result
> (illegal instruction) across different hardware.
>> 
>> Is this expected behavior? If so will the binaries of version 5.* have
> the same portability problem?
> 
> In short, yes, and yes. While discussions like
> http://redmine.gromacs.org/issues/1165 have taken place, nobody has decided
> to commit time to solving this, because the only benefit comes when the
> target machine architecture is variable, unknown, and there is no local
> mechanism in place to call the correct mdrun. This painful situation is
> partly created by choosing to have a  heterogeneous cluster... In 5.0 you
> will be able to compile for SSE4.1 and avoid RDTSC with the distributable
> build option, but this is wasting the capabilities of the AMD processors. A
> second GROMACS install, together with a recommended run script that chooses
> the mdrun binary based on the node name / whatever should be a reasonable
> solution.
> 
> Mark
> 
>> One more thing the situation described above occurs with both the gnu
> 4.4.7 compiler and the intel 13.1.1.
>> 
>> Any feedback would be greatly appreciated.
>> 
>> Thanks,
>> 
>> Raffaella.
>> 
>> --
>> 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.