[gmx-users] Gromacs on GPU

Téletchéa Stéphane stephane.teletchea at univ-nantes.fr
Fri Jun 16 20:27:38 CEST 2017


Le 16/06/2017 à 20:07, Mohsen Ramezanpour a écrit :
> Thanks Justin.
> 
> So, can we say that simulation on CPU and GPU (as far as we use the same
> version of Gromacs) are compatible?
> 
> If yes, is that okay to continue a simulation which was done with CPU (say
> till 100 ns) to 500 ns (using GPU)?
> Or I should start from t=0 with GPU?
> 
> It is important for my case as the allocations on supercomputers change
> from CPU to GPU. So, not sure if I should start all again or it is fine to
> continue.
> 
> Cheers

Dear Moshen,

What I'm doing daily is using my "small" workstation with CPU then 
continue once equilibration and primary production has worked well.

Go for GPU for a boost in performance (big) once the primary steps are 
ok :-)

Best,

Stéphane

-- 
Assistant Professor in BioInformatics, UFIP, UMR 6286 CNRS, Team Protein 
Design In Silico
UFR Sciences et Techniques, 2, rue de la Houssinière, Bât. 25, 44322 
Nantes cedex 03, France
Tél : +33 251 125 636 / Fax : +33 251 125 632
http://www.ufip.univ-nantes.fr/ - http://www.steletch.org


More information about the gromacs.org_gmx-users mailing list