Performance Portability Strategies for Grid C++ Expression Templates
- Univ. of Edinburgh, Scotland (United Kingdom). Higgs Centre for Theoretical Physics
- NVIDIA Corporation, Santa Clara, CA (United States)
- Univ. of Utah, Salt Lake City, UT (United States)
- Brookhaven National Laboratory (BNL), Upton, NY (United States)
One of the key requirements for the Lattice QCD Application Development as part of the US Exascale Computing Project is performance portability across multiple architectures. Using the Grid C++ expression template as a starting point, we report on the progress made with regards to the Grid GPU offloading strategies. We present both the successes and issues encountered in using CUDA, OpenACC and Just-In-Time compilation. Experimentation and performance on GPUs with a SU(3)×SU(3) streaming test will be reported. We will also report on the challenges of using current OpenMP 4.x for GPU offloading in the same code.
- Research Organization:
- Brookhaven National Laboratory (BNL), Upton, NY (United States)
- Sponsoring Organization:
- USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR)
- DOE Contract Number:
- SC0012704
- OSTI ID:
- 1424990
- Report Number(s):
- BNL--114831-2017-FORE
- Country of Publication:
- United States
- Language:
- English
Similar Records
Towards Achieving Performance Portability Using Directives for Accelerators
Automatic Offloading C++ Expression Templates to CUDA Enabled GPUs
Performance-Portable GPU Acceleration of the EFIT Tokamak Plasma Equilibrium Reconstruction Code
Conference
·
Tue Nov 01 00:00:00 EDT 2016
·
OSTI ID:1567436
Automatic Offloading C++ Expression Templates to CUDA Enabled GPUs
Conference
·
Tue May 01 00:00:00 EDT 2012
·
OSTI ID:1080421
Performance-Portable GPU Acceleration of the EFIT Tokamak Plasma Equilibrium Reconstruction Code
Conference
·
Sat Nov 11 23:00:00 EST 2023
· Proceedings of the SC '23 Workshops of The International Conference on High Performance Computing, Network, Storage, and Analysis
·
OSTI ID:2477210