[gmx-users] MD simulations on a remote workstation

Justin Lemkul jalemkul at vt.edu
Wed Apr 2 22:27:23 CEST 2014



On 4/2/14, 3:28 PM, Amjad Farooq wrote:
> Aldo,
>
> Thank you so very much for your prompt response.
>
> Well, I have my data files on the PC but MD simulations are running on the
> remote workstation. So, yes, you are right: it is the file write/transfer
> or I/O operation that will be affected.
>
> But, the question is by how much? How much is this gonna slow down my
> simulations? I wonder what the rate-limiting step in Gromacs is? Is it
> integrating the Newton's equation or writing data to disk?
>

I/O is expensive.  I/O over a network is even worse, but the impact on the 
simulation depends on how often you write to disk.  Is there some reason why you 
aren't writing to the disk on the remote workstation?  That eliminates any 
dependence on the interconnect completely.

-Justin

-- 
==================================================

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 601
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

==================================================


More information about the gromacs.org_gmx-users mailing list