what would be a good explanation of PyFR having a good utilization of GPU acceleration?

[Junting Chen]

Hello all,

I am trying to find a proper reason for the statement of "PyFR has a good utilization of GPU acceleration technology" , or possibly refer to a paper. I understand that memory transfer between CPU and GPU is one of the big slowdowns of GPU computing.

How would you say PyFR as a modern code uses GPU resource better than codes with histories then modified to utilize GPU acceleration? 

In the paper "pyfr an opensource frame work for solving advection-diffusion..." in 2014, it was said GEMM was optimized for large square matrices, where the constant operator in PyFR are small and square, and state matrices are short and fat. Is this improved?

Junting Chen

[Vincent, Peter E]

Hi Junting

I am trying to find a proper reason for the statement of "PyFR has a good utilization of GPU acceleration technology" , or possibly refer to a paper. I understand that memory transfer between CPU and GPU is one of the big slowdowns of GPU computing.

Here is an example of a paper that looks at performance aspects:

https://ieeexplore.ieee.org/document/7876999 

In the paper "pyfr an opensource frame work for solving advection-diffusion..." in 2014, it was said GEMM was optimized for large square matrices, where the constant operator in PyFR are small and square, and state matrices are short and fat. Is this improved?

GEMM can actually perform well in a range of scenarios. However, we have also developed technology for smaller/sparse matrices:

https://www.sciencedirect.com/science/article/pii/S0010465515004506

Peter

Dr Peter Vincent MSci ARCS DIC PhD FRAeS
Reader in Aeronautics and EPSRC Fellow
Department of Aeronautics
Imperial College London
South Kensington
London
SW7 2AZ
UK

--
You received this message because you are subscribed to the Google Groups "PyFR Mailing List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pyfrmailingli...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/pyfrmailinglist/e46b4ba3-67b2-4cc0-865d-b15596d0f430%40googlegroups.com.