On Wed, May 30, 2012, somedatta pal wrote:
> We installed AMBER 9 in our cluster computer. CPU model
> number of our cluster is Intel (R) Xeon (R) CPU X5472 . 300 GHz. The
> machine contains Red Hat Enterprise Linux 5.2 . Simulation of pure
> water (TIP3P model) is running fine in that machine and output files
> are also generated properly. But when we are running simulation of
> bio-molecules , it shows that sander.MPI is running but nothing is
> being written in the output files.
> AMBER 9 is also installed in other machines in our lab. In
> those machines simulation of bio-molecules, using same input files, is
> running fine and output files are also generated.
There is not nearly enough information provided for an outsider to provide
any assistance. Look at the simple test cases (e.g. dhfr) in parallel. If
they fail, then look carefully at the outputs for any clues as to what is
happening. Look for differences between the 4096wat test case and the dhfr
test case -- by the logic given above, the former should pass and the latter
should fail.
On the other hand, if the test cases pass, you will need to look for
differences between your jobs and those in the test cases. In any case, any
sample calculation you run while investigating this should set ntpr=1 and
nstlim to a small value (like 10 or 20).
....dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed May 30 2012 - 05:00:03 PDT
