Title: Evaluation of directive-based performance portable programming models

Authors: M. Graham Lopez; Wayne Joubert; Verónica G. Vergara Larrea; Oscar Hernandez; Azzam Haidar; Stanimire Tomov; Jack Dongarra

Addresses: Computational and Computer Sciences Directorate, Oak Ridge National Laboratory, Oak Ridge, TN, USA ' Computational and Computer Sciences Directorate, Oak Ridge National Laboratory, Oak Ridge, TN, USA ' Computational and Computer Sciences Directorate, Oak Ridge National Laboratory, Oak Ridge, TN, USA ' Computational and Computer Sciences Directorate, Oak Ridge National Laboratory, Oak Ridge, TN, USA ' Innovative Computing Laboratory, University of Tennessee, Knoxville, TN, USA ' Innovative Computing Laboratory, University of Tennessee, Knoxville, TN, USA ' Innovative Computing Laboratory, University of Tennessee, Knoxville, TN, USA

Abstract: We present an extended exploration of the performance portability of directives provided by OpenMP 4 and OpenACC to program various types of node architectures with attached accelerators. To do this, we use examples of algorithms with varying computational intensities for our evaluation, as both compute and data access efficiency are important considerations for overall application performance. We implement the kernels of interest using various methods provided by newer OpenACC and OpenMP implementations, and we evaluate their performance on various platforms including both x86\_64 and Power8 with attached NVIDIA GPUs, X86\_64 multicores, self-hosted Intel Xeon Phi KNL, as well as an X86\_64 host system with Intel Xeon Phi coprocessors. Furthermore, we present in detail what factors affected the performance portability, including how to pick the right programming model, its programming style, its availability on different platforms, and how well compilers can optimise and target multiple platforms.

Keywords: OpenMP 4; OpenACC; performance portability; programming models.

DOI: 10.1504/IJHPCN.2019.101254

International Journal of High Performance Computing and Networking, 2019 Vol.14 No.2, pp.165 - 182

Received: 23 Feb 2017
Accepted: 17 Apr 2017

Published online: 24 Jul 2019 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article