AMBER Archive (2008)

Subject: RE: AMBER: mpd error in submitting parallel job

From: Ross Walker (ross_at_rosswalker.co.uk)
Date: Thu Feb 14 2008 - 22:11:23 CST

> I'm not sure how it works for your machine, but in some configurations
> you may need to start the mpd in your script, such as:
> ----------
> mpdboot -v -f $PBS_NODEFILE

> > /nas/lpeng/src/amber9/exe/sander -O -i pgga10.in -p pgga10.top -c

You also need to use sander.MPI for parallel runs. You should also realize
that if this is a gigabit ethernet cluster rather than a 'real' interconnect
like infiniband you will likely be running much slower on 16 cpus that you
would on say 4. You should benchmark here to be sure. You might also want to
consider building PMEMD which if what you are simulating falls within what
it can do will likely be much faster in parallel.

All the best
Ross

/\
\/
|\oss Walker

| Assistant Research Professor |
| San Diego Supercomputer Center |
| Tel: +1 858 822 0854 | EMail:- ross_at_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.

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu