[gmx-users] mpich/share/machines file for dual processor machine

[Omololu Akin-Ojo]

Hi again,

i also use mpich (in my case i
use/opt/mpich.gcc/bin/mpirun). i have a file called
"mach" that contains the IP address of my dual
processor machine twice. It's content is:
172.16.100.1
172.16.100.1

(it would be different for you.
you can ping your machine -- from the same machine--to
get your address.)

then i just do:
mpirun -np 2 -machinefile mach
~/GROMACS/x86_64-unknown-linux-gnu/bin/mdrun -s
temp2.tpr -np 2 -v >&Uit1

and this works for me. Notice that i had to specify
the full path of my mdrun (in your case, mdrun_mpi)
command. Notice also how i included the file "mach".

o.




--- Jeroen Mesters <mesters at biochem.uni-luebeck.de>
wrote:

> Hi David,
> 
> 1) during the make step it used *mpicc* instead of
> normal cc. The latter 
> compiler was used when I left out the --enable-mpi.
> So, I think it is 
> really compiled for mpi??!? See also my comments
> further below.
> 2) I used either *mpirun -np 2 mdrun_mpi -s new.trp*
> or *mpirun -np 2 
> mdrun_mpi -s new.trp -np 2* and both produce the
> same first line in the 
> md.log file.
> 3) Installing from the gromacs home page lamxxx.rpm
> and fftwxxx.rpms 
> does not work properly on SuSE linux 9.2. Complains
> about
> 
> So my question is again, where does mdrun_mpi get
> the nnodes=1 
> information from?
> And, does anybody have an example of a machines file
> for dual processor!?
> 
> De groeten uit Luebeck! (Dutch for greetings from
> Luebeck!), Jeroen
> 
> David wrote:
> 
> >On Thu, 2005-09-29 at 17:06 +0200, Jeroen Mesters
> wrote:
> >  
> >
> >>Looking at the md.log file, the first line states
> the following: Log
> >>file opened: nodeid 0, nnodes = 1, host = orange,
> process = 7356
> >>
> >>Where does mdrun_mpi get the nnodes=1 information
> from?
> >>
> >>Does anyone have an mpich/share/machines file that
> shows me how a
> >>dual-processor machine is configured i.e. how do I
> tell mpich/gromacs
> >>there are two processors in one box and it should
> use both to run the
> >>job?
> >>    
> >>
> >
> >you sure that mdrun was compiled for mpi?
> >
> >try 
> >ldd `which mdrun_mpi` and see whether anyhing mpi
> shows up. If so then
> >check other commands to start mpi jobs like mpiexec
> or mpirun
> >
> >in general try to avoid mpich.
> >
> >  
> >
> >>Anybody?
> >>
> >>Thanx in advance, Jeroen.
> >>
> >>
> >>Jeroen Mesters wrote: 
> >>    
> >>
> >>>Machine: dual processor dell precision 650 with
> SuSE linux 9.2
> >>>Software: 1) mpich installed, machines file
> constructed, mpd started
> >>>2) compiled gromacs twice, once with and once
> without --enable-mpi
> >>>(using mpich). So, I ended up with two
> programmes, mdrun_mpi and
> >>>mdrun. The program versions are different in size
> which I would
> >>>expect... Indeed, one program was compiled with
> cc and the other
> >>>with mpicc (no errors during compilation).
> >>>
> >>>Step 1: grompp -np 2 -f full.mdp -c
> after_pr_CA.gro -p ph5.top -n
> >>>index.ndx -o 2cpu.tpr
> >>>Step 2: mdrun_mpi -np 2 -s 2cpu.tpr  >>>> Fatal
> error: run input
> >>>file 2cpu.tpr was made for 2 nodes, while
> mdrun_mpi expected it to
> >>>be for 1 nodes.
> >>>
> >>>Question 1. Are my commands correct i.e. is this
> the proper why to
> >>>start?
> >>>Question 2. Can this work at all or do I really
> need 2 computers (2
> >>>nodes)? 
> >>>
> >>>
> >>>Thanx for any hints, Jeroen.
> >>>-- 
> >>>Jeroen Raymundus Mesters, Ph.D.
> >>>Institut fuer Biochemie, Universitaet zu Luebeck
> >>>Ratzeburger Allee 160, D-23538 Luebeck
> >>>Tel: +49-451-5004070, Fax: +49-451-5004068
> >>>E-mail: mesters at biochem.uni-luebeck.de
> >>>Http://www.biochem.uni-luebeck.de
> >>>--
> >>>If you can look into the seeds of time and say
> >>>which grain will grow and which will not - speak
> then to me  (Macbeth)
> >>>--
> >>>  
> >>>
>
>>>____________________________________________________________________
> >>>
> >>>_______________________________________________
> >>>gmx-users mailing list
> >>>gmx-users at gromacs.org
> >>>http://www.gromacs.org/mailman/listinfo/gmx-users
> >>>Please don't post (un)subscribe requests to the
> list. Use the 
> >>>www interface or send it to
> gmx-users-request at gromacs.org.
> >>>      
> >>>
> >>-- 
> >>Jeroen Raymundus Mesters, Ph.D.
> >>Institut fuer Biochemie, Universitaet zu Luebeck
> >>Ratzeburger Allee 160, D-23538 Luebeck
> >>Tel: +49-451-5004070, Fax: +49-451-5004068
> >>E-mail: mesters at biochem.uni-luebeck.de
> >>Http://www.biochem.uni-luebeck.de
> >>--
> >>If you can look into the seeds of time and say
> >>which grain will grow and which will not - speak
> then to me  (Macbeth)
> >>--
> >>_______________________________________________
> >>gmx-users mailing list
> >>gmx-users at gromacs.org
> >>http://www.gromacs.org/mailman/listinfo/gmx-users
> >>Please don't post (un)subscribe requests to the
> list. Use the 
> >>www interface or send it to
> gmx-users-request at gromacs.org.
> >>    
> >>
> 
> 
> -- 
> Jeroen Raymundus Mesters, Ph.D.
> Institut fuer Biochemie, Universitaet zu Luebeck
> Ratzeburger Allee 160, D-23538 Luebeck
> Tel: +49-451-5004070, Fax: +49-451-5004068
> E-mail: mesters at biochem.uni-luebeck.de
> Http://www.biochem.uni-luebeck.de
> --
> If you can look into the seeds of time and say
> which grain will grow and which will not - speak
> then to me  (Macbeth)
> --
> 
> > _______________________________________________
> gmx-users mailing list
> gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please don't post (un)subscribe requests to the
> list. Use the 
> www interface or send it to
gmx-users-request at gromacs.org.



		
__________________________________ 
Yahoo! Mail - PC Magazine Editors' Choice 2005 
http://mail.yahoo.com