On the Performance Portability of Structured Grid Codes on Many-Core Computer Architectures

Simon McIntosh-Smith, Michael Boulton, Dan Curran, James Price
University of Bristol
Supercomputing: Lecture Notes in Computer Science, Volume 8488, pp 53-75, 2014





   series={Lecture Notes in Computer Science},

   editor={Kunkel, Julian Martin and Ludwig, Thomas and Meuer, Hans Werner},


   title={On the Performance Portability of Structured Grid Codes on Many-Core Computer Architectures},


   publisher={Springer International Publishing},

   keywords={Many-core; heterogeneous; GPU; Xeon Phi; structured grid; multi-grid multi-block; lattice Boltzmann},

   author={McIntosh-Smith, Simon and Boulton, Michael and Curran, Dan and Price, James},




Download Download (PDF)   View View   Source Source   



With the advent of many-core computer architectures such as GPGPUs from NVIDIA and AMD, and more recently Intel’s Xeon Phi, ensuring performance portability of HPC codes is potentially becoming more complex. In this work we have focused on one important application area — structured grid codes — and investigated techniques for ensuring performance portability across a diverse range of different, high-end many-core architectures. We chose three codes to investigate: a 3D lattice Boltzmann code (D3Q19 BGK), the CloverLeaf hydrodynamics mini application from Sandia’s Mantevo benchmark suite, and ROTORSIM, a production-quality structured grid, multiblock, compressible finite-volume CFD code. We have developed OpenCL versions of these codes in order to provide cross-platform functional portability, and compared the performance of the OpenCL versions of these structured grid codes to optimized versions on each platform, including hybrid OpenMP/MPI/AVX versions on CPUs and Xeon Phi, and CUDA versions on NVIDIA GPUs. Our results show that, contrary to conventional wisdom, using OpenCL it is possible to achieve a high degree of performance portability, at least for structured grid applications, using a set of straightforward techniques. The performance portable code in OpenCL is also highly competitive with the best performance using the native parallel programming models on each platform.
Rating: 2.5. From 1 vote.
Please wait...

* * *

* * *

HGPU group © 2010-2017 hgpu.org

All rights belong to the respective authors

Contact us: