Analysis of an Efficiency of Parallelization of Algorithms Running on Computing Cluster Based on Single-Board Diskless Computers Raspberry PI 3 Model B

Autor:	Victor V. Makarov, Darya V. Matveeva, Aleksandr B. Vavrenyuk, Nikita M. Lukyantsev
Rok vydání:	2021
Předmět:	Debugging Parallel processing (DSP implementation) Computer science Process (engineering) media_common.quotation_subject Computer cluster Linear system Parallel computing Cluster (spacecraft) Cluster analysis Queue Algorithm media_common
Zdroj:	2021 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering (ElConRus).
DOI:	10.1109/elconrus51938.2021.9396277
Popis:	This article is devoted to the analysis of parallelization efficiency of algorithms for solving linear systems of equations. The article presents an assessment of the quality of optimization of two algorithms for distributing the computational load between cluster nodes. The hardware is a computing cluster based on single-board computers with Linux Debian 8 as an operation system. Hydra Process Manager with an Slurm resource manager that implements a queue for multi-user access to the system is used to run and debug parallel tasks. The resulting system is planned to be used in the future in the educational process of MEPhI (Moscow Engineering Physics Institute) for a laboratory workshop on parallel computing.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::32b39f3750b7a83141ffcee3730dc257 https://doi.org/10.1109/elconrus51938.2021.9396277 Zobrazit plný text záznamu