GSOC 2013 Parallelized Backend
Hello, This is Junaid form National University of Sciences and Technology. I have gone through the GSOC 2013 idea listing. I am pretty much interested in extension of odeint. I am more interested for the parallelization over the GPUs preferably NVIDIA. From the details listed there, it seems that Boost community is more interested in parallelization over the CPUs i.e. much preference is given to OpenMP and MPI. Is that the case here? Regards, Junaid.
Hi Junaid, On 04/15/2013 05:53 PM, Muhammad Junaid Muzammil wrote:
Hello,
This is Junaid form National University of Sciences and Technology. I have gone through the GSOC 2013 idea listing. I am pretty much interested in extension of odeint. I am more interested for the parallelization over the GPUs preferably NVIDIA. From the details listed there, it seems that Boost community is more interested in parallelization over the CPUs i.e. much preference is given to OpenMP and MPI.
odeint already provides parallelization over GPUs via Thrust, VexCL and ViennaCL. So, there is no specific need for separate GPU backends. But there is currently no backend for MPI. Best regards, Karsten
Is that the case here?
Regards, Junaid.
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
participants (2)
-
Karsten Ahnert
-
Muhammad Junaid Muzammil