[gmx-users] scalability of gromacs-3.3

Anupam Nath Jha anupam at mbu.iisc.ernet.in
Wed Dec 26 06:25:39 CET 2007

Dear all
 I am trying to scale gromacs on BG/L and I am able to run mdrun on BG/L for
32processors. but when i try to run for 64 or 96, by making the *.tpr file:

grompp -np 96 -f run.mdp -p topol.top -c pr.gro -o run.tpr

and submitting in BG/L:
using the following submission file---

# @ job_type = bluegene
# @ executable = /usr/bin/mpirun
# @ arguments = -exe /home/phd/04/mbuanjha/soft/gmx332/bin/mdrun_mpi -cwd
/gpfs/bglscratch/mbuanjha/test/ -args "-s rum.tpr -o run.trr -c run.gro -e
run.edr -g test.log -v"
# @ error = test.$(host).$(cluster).$(process).err
# @ output = test.$(host).$(cluster).$(process).out
# @ environment = COPY_ALL
# @ wall_clock_limit=100:00:00
# @ notification = always
# @ bg_size = 96
# @ class = high
# @ queue

i am getting this error--

Getting Loaded...
Reading file rum.tpr, VERSION 3.3.2 (single precision)

Program mdrun_mpi, VERSION 3.3.2
Source code file: init.c, line: 69

Fatal error:
run input file rum.tpr was made for 96 nodes,
             while mdrun_mpi expected it to be for 128 nodes.

"Carry Me Away" (Motors)

Error on node 0, will try to stop all the nodes
Halting parallel program mdrun_mpi on CPU 0 out of 128

is it something like that gromacs can't run for 64 or 96nodes or what.

not only this when i try to submit for 128nodes, it's giving some memory
corruption error,  like:

*** *** glibc detected *** malloc(): memory corruption***

can you help me out?

with best regards

Science is facts; just as houses are made of stone, so is science is made of
facts; but a pile of stones is not a house, and  a collection of facts is not
necessarily science.

Anupam Nath Jha
Ph. D. Student
Saraswathi Vishveshwara Lab
Molecular Biophysics Unit
Ph. no.-22932611

This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.

More information about the gromacs.org_gmx-users mailing list