[gmx-users] Re: Parallel Use of gromacs - parallel run is 20 times SLOWER than single node run (Jim Kress)

Erik Lindahl lindahl at cbr.su.se
Wed Jul 18 18:51:06 CEST 2007

Hi Jim,

That system should scale quite well with a fast network (infiniband)  
since you don't seem to be using PME, and the CVS 3.99 version is  
even better.

However, as David already mentioned, the gigabit network is likely  
limiting you. If you test the CVS code there are timing reports that  
will tell you how much _time_ (vs. floating-point operations) is  
spent on each part of the algorithm.



On Jul 18, 2007, at 6:23 PM, Jim Kress wrote:

> It's a 3.712 cube containing 1722 tip4p water molecules at 300 deg  
> K and
> density = 1000kg/l
> Regards,
> Jim Kress
> _______________________________________________
> gmx-users mailing list    gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please search the archive at http://www.gromacs.org/search before  
> posting!
> Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-request at gromacs.org.
> Can't post? Read http://www.gromacs.org/mailing_lists/users.php

More information about the gromacs.org_gmx-users mailing list