Index  | Recent Threads  | Unanswered Threads  | Who's Active  | Guidelines  | Search
 

Quick Go »
No member browsing this thread
Thread Status: Active
Total posts in this thread: 3
[ Jump to Last Post ]
Post new Thread
Author
Previous Thread This topic has been viewed 1527 times and has 2 replies Next Thread
Loui_h20
Cruncher
Joined: Aug 12, 2013
Post Count: 25
Status: Offline
Reply to this Post  Reply with Quote 
Dr Gan and the Nvidia collaboration

Just doing a bit of digging and found this "Dr Gan joined Q-Chem as a staff scientist in 2006. His major responsiblility includes Q-Chem code development, maintenance, platform porting and code releasing. He also provides technical support and consultations to Q-Chem customers and developers around the world. His research project in Q-Chem includes the development of new DFT functionals. and high performance algorithms in computing DFT exchange Fock matrix. He also works with industrial projects. He has ported Q-Chem to SUN computing grid (network.com), and is collaborating with NVidia and Prof. Alán Aspuru-Guzik's group at Harvard University to make use of GPU acceleration in Q-Chem. "

Is Dr Gan still working on it? Is the lab still in touch with Dr Gan?
----------------------------------------

[Dec 19, 2013 10:45:17 AM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Dr Gan and the Nvidia collaboration

There are complex non-disclosure agreements in place between WCG[IBM] and Q-Chem, it being a commercial product, so asking questions and getting answers could be a sensitive matter. WCG does have their source code [it would not have gotten on the grid otherwise] and has been working with the involved parties to better the engine... there's some serious talented people in the WCG team who have an accumulated experience of many years to griddify applications [and let them run multiple times faster as they did with CFSW and HCC].

As for GPGPU coming to CEP2... there's absolutely been no noise when holding ear to track, there's just a CEP2v2 underway, which will bring improvements. I'm hoping for one on the same method of task set-up as was first time done with MCM... symbolic linking to application/reference files rather than copying to each job slot those 6700 work static files each CEP2 job slot has. We'll see.
[Dec 19, 2013 11:21:04 AM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Dr Gan and the Nvidia collaboration

Yes, Zhengting is still one of the staff scientists and programmers at Q-Chem and also one of our go-to-people over there. He and some of the other Q-Chem guys (in particular Yihan Zhao and Alex Sodt) have been a tremendous help over the years.
As for the use of GPUs: Yes, it is still on our wishlist, BUT considering our very limited manpower, we have decided to set our priorities differently for the time being. We still hope that we'll get around to this at some point in the future but as of now there are no concrete plans.
Best wishes,
Your CEP team
[Jan 21, 2014 9:21:59 PM]   Link   Report threatening or abusive post: please login first  Go to top 
[ Jump to Last Post ]
Post new Thread