Conveners
Parallel session 2B : Parallelization Efforts
- John Apostolakis (CERN)
Description
The focus of this session is on particular topics in Parallelisation:
- Geant4MT: open/new requirements and the potential for porting to new threads;
- Novel parallelization efforts including GPU porting and vectorization.
Notes:
1) Time slots without a speaker are for discussions.
2) The timetable is in nearly final version. Please notify in case of errors or issues.
This session will use the EVO system. The Id of the meeting is 5782381.
To join you can either
a) call one of the EVO phone bridge phone numbers near you. After the meeting Id you will be asked for a second number - enter "3296". The phone bridge numbers can be found at http://evo.caltech.edu/evoGate/telephone.jsp
b) use Skype and the EVO-Skype bridge - you need to use the same meeting id and use the same 2nd number "3296". Instructions are available at
http://evo.caltech.edu/evoGate/Documentation/extclient/skype/skype.html
c) use the Evo client - best only if you are familiar with it. To do this follow "Video Services" in the tab on the left and follow the link to join. The number you need is "1984".
-
Dr John Apostolakis (CERN)10/09/2012, 16:00
-
Dr Marc Paterno (Fermilab)10/09/2012, 16:05Perspectives on the potential benefits and issues with porting Geant4-MT to C++ 11 threads.Go to contribution page
-
10/09/2012, 16:20How to address the requirement for co-working with TBB (dispatch model of parallelism) ? - Analysis of requirement - Discussion of impact - Conclusions and actionsGo to contribution page
-
Dr John Apostolakis (CERN)10/09/2012, 16:35Introduction / contextGo to contribution page
-
Dhruva Tirumala Bukkapatnam10/09/2012, 16:40
-
Mr Philippe Canal (Fermi National Accelerator Lab. (US)), Dr Soon Yung Jun (Fermi National Accelerator Lab. (US))10/09/2012, 16:50Results from prototypes with GPUs, Vectors. Information on hardware evolution, including MIC.Go to contribution page
-
Koichi Murakami (High Energy Accelerator Research Organization (JP))10/09/2012, 17:20
-
10/09/2012, 17:35
-
Andrei Gheata (CERN)10/09/2012, 17:45