Empirical investigation: performance and power‐consumption based dual‐level model for exascale computing systems. Issue 4 (1st August 2020)

• Record Type: Journal Article
• Authors: Ashraf, Muhammad Usman; Eassa, Fathy Alboraei; Ahmad, Aiiad; Algarni, Abdullah
• Is Part Of: IET software. Volume 14:Issue 4(2020)
• Page Start: 319
• Page End: 327
• Publication Date: 2020-08-01
• Subjects: message passing -- application program interfaces -- parallel programming -- iterative methods -- multiprocessing systems -- matrix multiplication -- power aware computing; power-consumption -- dual-level model -- exascale computing systems -- Exaflop speed -- power consumption -- ultrascale performance -- thousand-fold enhancement -- high-performance computing -- massive parallelism -- promising parallel programming approach -- ExaFlops -- hybrid parallel programming models -- homogeneous cluster systems -- heterogeneous cluster systems -- hybrid models -- homogeneous parallel programming model -- promising model -- HPC -- power 20.0 MW; Computer software -- Periodicals; Software engineering -- Periodicals; 005.1
• DOI: 10.1049/iet-sen.2018.5062
• Languages: English
• ISSNs: 1751-8806

Abstract

Abstract : Exascale computing systems (ECS) are anticipated to perform at Exaflop speed (10 18 operations per second) using power consumption <20 MW. This ultrascale performance requires the speedup in the system by thousand‐fold enhancement in current Petascale. For future high‐performance computing (HPC), power consumption is one of the vital challenges faced to achieve Exaflops through the traditional way of increasing clock‐speed. One standard way to attain such significant performance is through massive parallelism. In the early stages, it is hard to decide the promising parallel programming approach that can provide massive parallelism to attain ExaFlops. This article commences with a short description and implementation of algorithms of various hybrid parallel programming models (PPMs) for homogeneous and heterogeneous cluster systems. Furthermore, the authors evaluated performance and power consumption in these hybrid models by implementing in two HPC benchmarking applications such as square matrix multiplication and Jacobi iterative solver for two‐dimensional Laplace equation. The results demonstrated that the hybrid of heterogeneous (MPI + X ) outperformed to homogeneous parallel programming (MPI + OpenMP) model. This empirical investigation of hybrid PPMs is a leading step for researchers and development communities to select a promising model for emerging ECS.