[gmx-users] Multiprocessor problems

[Sampo Karkola]

Quoting David <spoel at xray.bmc.uu.se>:

> On Tue, 2003-06-17 at 09:17, Sampo Karkola wrote:
> > Dear users,
> > 
> > I'm running md simulations for a protein in water with ions and ran
> into
> > some problems when I started using a multiprocessor computer. I have
> several
> > runs with and without substrate or water molecules in the active site.
> I was
> > able to energy minimise and run a restricted dynamics simulation on
> them
> > with no problems with the multiprocessor machine but when I tried the
> free
> > dynamics I encountered several problems.
> > 
> > One mdrun doesn't start because it complains about an iron-coordinated
> > sulfur having more than 10 bonds. I checked the structure file but
> > everything seems to be ok. I tried the mdrun on a single processor
> machine
> > and it worked fine. The only difference between the mdruns was the
> number of
> > nodes defined during grompp. I noticed that someone else had a similar
> > problem earlier in the list but as far as I searched, there were no
> answers
> > to him. So, if someone knows how to fix this, I'd be grateful.
> > 
> > Secondly, again in multiprocessor machine, one mdrun doesn't start and
> I get
> > an error message saying "Format exec error. Binary file not
> executable".
> > Again, this works fine in a single processor mode and the only
> difference is
> > the number of nodes defined during grompp.
> 
> Please supply more information on gromacs version, machine type number
> of processors and simulation system.
> 
> If you haven't got the latest gmx version (3.1.4) that would a good
> start...
> -- 

I'm using gromacs3.0.3. and updating it is out of my hands. The machine is
an IBM eServer Cluster 1600 with 16 IBM p690 nodes, each equipped with 32
Power4 processors. The simulation consists of a 500 aa protein with a heme
group with/without a substrate and water or oxygen in the active site. There
are about 10000 water molecules and around 70 ions. The first mdrun was made
with the backbones of the SCRs fixed and the second would be made with a
free backbone.

TIA,

Sampo



> Groeten, David.
> ________________________________________________________________________
> Dr. David van der Spoel, 	Dept. of Cell and Molecular Biology
> Husargatan 3, Box 596,  	75124 Uppsala, Sweden
> phone:	46 18 471 4205		fax: 46 18 511 755
> spoel at xray.bmc.uu.se	spoel at gromacs.org   http://xray.bmc.uu.se/~spoel
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> _______________________________________________
> 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.
> 


*********************************************
Sampo Karkola, M.Sc.
Scientific researcher, Teaching assistant
Laboratory of Organic Chemistry
Department of Chemistry
PO Box 55
FIN-00014 University of Helsinki
Finland

e-mail: Sampo.Karkola at helsinki.fi
http://www.physics.helsinki.fi/~karkola
tel:    +358 9 19150369 Work
        +358 50 5871175 Mobile
fax:    +358 9 19150366