AMBER Archive (2009)

Subject: Re: [AMBER] RE: pmemd parallel

From: Jun Dong (jun_at_strubi.ox.ac.uk)
Date: Wed Jul 15 2009 - 10:00:49 CDT


Dear Ross,
Many thanks for all your help. I cannot find anything useful in 4096 test directory. Test failed and no errors show in the 4096 test directory. We want to run amber on a computer which has 2 quad core processors so no ssh or rsh is required. I believe I can run pmemd parallel as
mpiexec -np 8 pmemd
I will need to read the manual before I will find out if it works.
Best regards,
Jun Dong
Division of Structural Biology
The Wellcome Trust Centre for Human Genetics
Oxford University
Roosevelt Drive
Oxford
OX3 7BN
Tel: 01865 287558

---- Original message ----
>Date: Tue, 14 Jul 2009 05:24:06 -0700
>From: amber-bounces_at_ambermd.org (on behalf of "Ross Walker" <ross_at_rosswalker.co.uk>)
>Subject: [AMBER] RE: pmemd parallel
>To: <jun_at_strubi.ox.ac.uk>
>Cc: "'AMBER Mailing List'" <amber_at_ambermd.org>
>
>Hi Jun,
>
>> Dear Ross,
>> I did test for pmemd parallel. We want to run pmemd parallel on a 8
>> processor linux computer.
>>
>> amber10/test> setenv DO_PARALLEL "mpiexec -n 8"
>> amber10/test> make test.pmemd
>> export TESTsander='../../exe/pmemd'; cd 4096wat && ./Run.pure_wat
>> diffing mdout.pure_wat.save with mdout.pure_wat
>> possible FAILURE: check mdout.pure_wat.dif
>> ==============================================================
>> export TESTsander='../../exe/pmemd'; cd 4096wat &&
>> ./Run.pure_wat_nmr_temp_reg
>> diffing mdout.pure_wat_nmr_temp.save with mdout.pure_wat_nmr_temp
>> possible FAILURE: check mdout.pure_wat_nmr_temp.dif
>> ==============================================================
>...
>...
>> How to run pmemd parallel on a 8 processor machine? Is there a problem
>> for our pmemd?
>
>This should work fine. Are you sure your DO_PARALLEL command is correct for
>your mpi? Did you check the dif files to see what they say? I suspect there
>is an error such as some mpi demon is not running or you also need to
>provide a node list, or you don't have passwordless ssh/rsh access etc.
>Taking a look inside the 4096wat test directory at mdout.pure_wat and
>mdout.pure_wat.dif would be a good start to find out what is going wrong.
>
>Also check you can run some MPI tests - most mpi installations come with
>their own test directory.
>
>All the best
>Ross
>
>/\
>\/
>|\oss Walker
>
>| Assistant Research Professor |
>| San Diego Supercomputer Center |
>| Tel: +1 858 822 0854 | EMail:- ross_at_rosswalker.co.uk |
>| http://www.rosswalker.co.uk | PGP Key available on request |
>
>Note: Electronic Mail is not secure, has no guarantee of delivery, may not
>be read every day, and should not be used for urgent or sensitive issues.
>
>
>
>
>
>_______________________________________________
>AMBER mailing list
>AMBER_at_ambermd.org
>http://lists.ambermd.org/mailman/listinfo/amber

_______________________________________________
AMBER mailing list
AMBER_at_ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber