Re: [gmx-users] MPICH or LAM/MPI

Arneh Babakhani wrote: Hi Carsten, thanks for the reply, good question. I can run it fine on as much as 4 processors, but nothing beyond that. Any idea why? Hi Arneh, which version of mpich are you using? Several people have encountered hangs on >4 processors when running GROMACS on top of

Re: [gmx-users] MPICH or LAM/MPI

Hi Carsten, thanks for the reply, good question. I can run it fine on as much as 4 processors, but nothing beyond that. Any idea why? Arneh Carsten Kutzner wrote: Hi Arneh, do you have the same problem on less processors? Can you run on 1, 2 and 4 procs? Carsten Arneh Babakhani wrote:

Re: [gmx-users] MPICH or LAM/MPI

Hi Arneh, do you have the same problem on less processors? Can you run on 1, 2 and 4 procs? Carsten Arneh Babakhani wrote: Hi All, Ok, I've successfully created the mpi version of mdrun. Am now trying to run my simulation on 32 processors. After processing with grompp and the option -np 32,

Re: [gmx-users] MPICH or LAM/MPI

Hi All, Ok, I've successfully created the mpi version of mdrun. Am now trying to run my simulation on 32 processors. After processing with grompp and the option -np 32, I use mdrun with the following script (where CONF is the input file, NPROC is the number of processors): /opt/mpich/intel/bin/

Re: [gmx-users] MPICH or LAM/MPI

Hi Carsten, Yep, that was exactly it. When I included that path in my LD_LIBRARY_PATH variable, it configured and compiled no problem. Thanks! Arneh Carsten Kutzner wrote: Hi Arneh, the problem seems to me that the program that has to determine the size of an int cannot be compiled due t

Re: [gmx-users] MPICH or LAM/MPI

Hi Arneh, the problem seems to me that the program that has to determine the size of an int cannot be compiled due to a missing LAM library: --- error while loading shared libraries: liblamf77mpi.so.0: cannot open shared object file: No such file or directory --- So make shure that this lib

Re: [gmx-users] MPICH or LAM/MPI

Hello, on this note, I'm having issues configuring and installing the mpi enabled mdrun. (see below). I keep getting a strange error: checking size of int... configure: error: cannot compute sizeof (int), 77 I've pasted the output below, and I've attached the config.log for reference.  I'm rea

Re: [gmx-users] MPICH or LAM/MPI

computer nodes. Carsten -Original Message- From: Carsten Kutzner <[EMAIL PROTECTED]> To: Discussion list for GROMACS users Sent: Mon, 19 Jun 2006 10:28:51 +0200 Subject: Re: [gmx-users] MPICH or LAM/MPI Hello Hector, since it does not take long to install lam and mpich, I would i

Re: [gmx-users] MPICH or LAM/MPI

<[EMAIL PROTECTED]> To: Discussion list for GROMACS users Sent: Mon, 19 Jun 2006 10:28:51 +0200 Subject: Re: [gmx-users] MPICH or LAM/MPI Hello Hector, since it does not take long to install lam and mpich, I would install both MPI implementations and then benchmark with a typical MD

Re: [gmx-users] MPICH or LAM/MPI

Hello Hector, since it does not take long to install lam and mpich, I would install both MPI implementations and then benchmark with a typical MD system which one performs better. I would suggest LAM 7.1.2 which is the newest version, and MPICH-2 1.0.3, which works well, at least on our clus

[gmx-users] MPICH or LAM/MPI

Dear Gmx-Users, I'm configuring a small cluster of 5 Dell Power Edge 1800 with dual xeon conected by a gigabit dell switch. I was wondering which MPI distribution, LAM/MPI or MPICH(which version), on this platform has better performance for gromacs 3.3.1 on a fedora 5 linux distribution. I w