Performance and Energy Analysis of OpenMP Runtime Systems with Dense Linear Algebra Algorithms

Abstract : In this paper, we analyse performance and energy consumption of five OpenMP runtime systems over a 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 DVFS 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. A LU factorization with concurrent write extension from libKOMP achieved up to 63% of performance gain and 29% of energy decrease.
Complete list of metadatas

Cited literature [14 references]  Display  Hide  Download

https://hal.inria.fr/hal-01957220
Contributor : Thierry Gautier <>
Submitted on : Monday, December 17, 2018 - 10:59:17 AM
Last modification on : Thursday, February 7, 2019 - 5:22:50 PM
Long-term archiving on : Monday, March 18, 2019 - 2:09:44 PM

File

ijhpca-wamca2017.pdf
Files produced by the author(s)

Identifiers

Collections

Citation

Joao Vicente Ferreira Lima, Issam Raïs, Laurent Lefèvre, Thierry Gautier. Performance and Energy Analysis of OpenMP Runtime Systems with Dense Linear Algebra Algorithms. International Journal of High Performance Computing Applications, SAGE Publications, 2018, pp.1-17. ⟨10.1177//1094342018792079⟩. ⟨hal-01957220⟩

Share

Metrics

Record views

69

Files downloads

142