Performance and energy analysis of OpenMP runtime systems with dense linear algebra algorithms. (May 2019)

• Record Type: Journal Article
• Authors: Ferreira Lima, João Vicente; Raïs, Issam; Lefèvre, Laurent; Gautier, Thierry
• Other Names: Mencagli Gabriele guest-editor.; França Felipe MG guest-editor.; Bentes Cristiana Barbosa guest-editor.; Justen Marzulo Leandro Augusto guest-editor.; Lima Pilla Mauricio guest-editor.; Wyrzykowski Roman guest-editor.; Deelman Ewa guest-editor.
• Is Part Of: International journal of high performance computing applications. Volume 33:Number 3(2019)
• Page Start: 431
• Page End: 443
• Publication Date: 2019-05
• Subjects: OpenMP -- task parallelism -- linear algebra algorithms -- NUMA -- energy efficiency; High performance computing -- Periodicals; Supercomputers -- Periodicals; 004.1105
• DOI: 10.1177/1094342018792079
• Languages: English
• ISSNs: 1094-3420

Abstract

In this article, we analyze performance and energy consumption of five OpenMP runtime systems over a non-uniform memory access (NUMA) platform. We also selected three CPU-level optimizations or techniques to evaluate their impact on the runtime systems: processors features Turbo Boost and C-States, and CPU Dynamic Voltage and Frequency Scaling through Linux CPUFreq governors. We present an experimental study to characterize OpenMP runtime systems on the three main kernels in dense linear algebra algorithms (Cholesky, LU, and QR) in terms of performance and energy consumption. Our experimental results suggest that OpenMP runtime systems can be considered as a new energy leverage, and Turbo Boost, as well as C-States, impacted significantly performance and energy. CPUFreq governors had more impact with Turbo Boost disabled, since both optimizations reduced performance due to CPU thermal limits. An LU factorization with concurrent-write extension from libKOMP achieved up to 63% of performance gain and 29% of energy decrease over original PLASMA algorithm using GNU C compiler (GCC) libGOMP runtime.