Parallel Programming Hybrid applied in a Cluster Computing with tests on Matrix Multiplication

  • Leonardo Mendes URI
  • Eduardo Ferreira URI
DOI: https://doi.org/10.5753/eradrs.2020.10763

Abstract


This article highlights the use of two parallel tools, MPI and OpenMP, applied in a computer cluster. The program takes place in hybrid form, and the MPI used to parallelize the exchange of cluster messages, and OpenMP to parallelize the number of each node processors. The tests are applied using a multiplication algorithm arrays, and the results are compared with a computer with twice the processing of the cluster nodes.

Keywords: Algorithms Parallel and Distributed, Languages, Compilers and Tools Parallel and Distributed Computing, Distributed systems

References

Alecrim, E. (2013) “Cluster: conceito e características”, https://www.infowester.com/cluster.php, Janeiro.

Bacellar, H. Viana. (2010) “Cluster: Computação de Alto Desempenho”, http://www.ic.unicamp.br/~ducatte/mo401/1s2010/T2/107077-t2.pdf, Janeiro.

Favaretto, R. M. (2018) “Uma análise de eficiência do uso das ferramentas de programação paralela OpenMP e TBB na remoção de ruídos em imagens”. Revista THEMA, vol, 15. Nº 4, pág. 1234 a 1255.

Marongiu, A., Burgio, P. e Benini, L. Supporting OpenMP on a multi-cluster embedded MPSoC. In: Microprocessors and Microsystems, 8. 2011, Amsterdam. Proceedings at Microprocess Microsyst. Amsterdam: Elsevier Science Publishers, 2011, p. 668-682.

MPI. (2012) “MPI: A Message-Passing Interface Standard Version 3.0”, Message Passing Interface Forum.
Published
2020-04-15
How to Cite

Select a Format
MENDES, Leonardo; FERREIRA, Eduardo. Parallel Programming Hybrid applied in a Cluster Computing with tests on Matrix Multiplication. In: REGIONAL SCHOOL OF HIGH PERFORMANCE COMPUTING FROM SOUTHERN BRAZIL (ERAD-RS), 20. , 2020, Santa Maria. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2020 . p. 89-92. ISSN 2595-4164. DOI: https://doi.org/10.5753/eradrs.2020.10763.