Hi Ilyas,
Try doing:
export DO_PARALLEL='mpirun -np 2'
Even though you build the parallel version on a single cpu machine you still
have to run more than 1 thread when doing thermodynamic integration. It just
runs both threads, time sliced, on the same processor so it does not run
twice as fast as it would on a dual processor. Make sure your machines.dat
file (or mpi installation equivalent) specifies the local single cpu machine
more than once so you can do things like mpirun -np 2 on a single cpu
machine.
All the best
Ross
/\
\/
|\oss Walker
| HPC Consultant and Staff Scientist |
| San Diego Supercomputer Center |
| Tel: +1 858 822 0854 | EMail:- ross.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.
> -----Original Message-----
> From: owner-amber-developers.scripps.edu
> [mailto:owner-amber-developers.scripps.edu] On Behalf Of
> Ilyas Yildirim
> Sent: Friday, June 09, 2006 19:28
> To: amber-developers.scripps.edu
> Subject: amber-developers: AMBER 9 in a single CPU
>
> Dear All,
>
> In the AMBER 9 Manual, pp. # 8, it says that we can build the parallel
> ver. of AMBER 9 for a machine with a single CPU. My linux box
> is a single
> CPU P4 machine (SUSE 9.1 operating system). I have installed
> the parallel
> version by following the installation protocols. When I try
> to test the
> test cases, it is giving the following error:
> --------------------------------------------------------------
> -----------
> ==============================================================
> cd bintraj; ./Run.bintraj
>
> sander and ptraj: test sander netCDF output and ptraj netCDF input
> [0] MPI Abort by user Aborting program !
> [0] Aborting program!
> p0_19949: p4_error: : 1
> ./Run.bintraj: Program error
> make[1]: [test.sander.BASIC] Error 1 (ignored)
> make[1]: Leaving directory `/programs/amber9/test'
> export TESTsander=/home/yildirim/amber9/exe/sander.MPI; cd 4096wat;
> ./Run.column_fft
> diffing mdout.column_fft.save with mdout.column_fft
> PASSED
> ==============================================================
> export TESTsander=/home/yildirim/amber9/exe/sander.MPI; cd jar_multi;
> ./Run.jar
> Error: specified more groups ( 2 ) than the number of
> processors (
> 1 ) !
> [0] MPI Abort by user Aborting program !
> [0] Aborting program!
> p0_20101: p4_error: : 1
> ./Run.jar: Program error
> make: *** [test.sander.BASIC.MPI] Error 1
> --------------------------------------------------------------
> -----------
> In the manual (the same page), it is saying that free energy
> calculations
> can be done if the parallel ver. of AMBER 9 is compiled in a
> single CPU
> machine. I saw in Ross Walker's following website that he compiled
> the single CPU sander.MPI for SDSC (Datastar):
>
> http://coffee.sdsc.edu/amber_sdsc/datastar.htm
>
> Maybe I am doing something wrong with the compilation, but during the
> parallel version compilation, I did not get any error messages. I will
> appreciate if someone can comment on the error message I am getting.
>
> Best,
>
> --
> Ilyas Yildirim
> ---------------------------------------------------------------
> - Department of Chemisty - -
> - University of Rochester - -
> - Hutchison Hall, # B10 - -
> - Rochester, NY 14627-0216 - Ph.:(585) 275 67 66
> (Office) -
> - http://www.pas.rochester.edu/~yildirim/ -
> ---------------------------------------------------------------
>
>
Received on Sun Jun 11 2006 - 06:07:19 PDT
