From: Axel Kohlmeyer (akohlmey_at_cmm.chem.upenn.edu)
Date: Mon Jun 15 2009 - 13:19:05 CDT
On Mon, 2009-06-15 at 10:44 -0700, Lixia Jin Day wrote:
> Dear all,
> We recently switched to our project to another computer cluster, which
> we have to pay 2.5 cents for every CPU time. As our system is huge,
> over 350,000 atoms, I need to find a better way to run our simulation.
> Please give some of your advices. Thank you!
1) have a close look at the NamdWiki. there are many useful suggestions
to optimize your parameters.
2) make some benchmark tests and test for maximum scaling and maximum
throughput. those are usually not the same and you can make a choice
as to how much money it is worth to you to get your results quickly.
p.s.: computer time on teragrid "only" costs writing a
-- ======================================================================= Axel Kohlmeyer akohlmey_at_cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot.
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:52:55 CST