Message: Parallelism in G4 on video-cards Not Logged In (login)
 Next-in-Thread Next-in-Thread
 Next-in-Forum Next-in-Forum

Question Parallelism in G4 on video-cards 

Forum: User Requirements
Date: 17 Sep, 2009
From: Dmitry Naumov <Dmitry Naumov>

Dear G4 developers and users. I would like to learn from you if anyone tried to run g4 simulations in parallel jobs using processors of video-cards? The point is that such kind of unusual use of video-cards becomes very powerfull in recent years and people get factors tens-hundreds of performance increase using this approach (see http://www.nvidia.com/object/cuda_home.html#)

I would like to know first of all from the developers if G4 design allows such a splitting, and which parts of g4 could be used for parallel jobs and where it is better not to spend time on this kind of developments.

Some of my colleagues are involved in doing physics simulation using nvidia cards parallelism and they volunteer to work in this direction for g4 if this was not done already, and if g4 design is suitable for this. Thank you, Dmitry

Inline Depth:
 1 1
 All All
Outline Depth:
 1 1
 2 2
 All All
Add message: (add)

1 None: Re: Parallelism in G4 on video-cards   (Paul Nicholas Colin Gloster - 18 Sep, 2009)
(_ None: Re: Parallelism in G4 on video-cards   (Joseph Perl - 18 Sep, 2009)
 Add Message Add Message
to: "Parallelism in G4 on video-cards"

 Subscribe Subscribe

This site runs SLAC HyperNews version 1.11-slac-98, derived from the original HyperNews


[ Geant 4 Home | Geant 4 HyperNews | Search | Request New Forum | Feedback ]