Uma Metodologia Baseada em Simulação e Algoritmo Genético para Exploração de Estratégias de Escalonamento de Laços

Abstract : In High Performance Computing, the application's workload must be well balanced among the threads to achieve better performance. In this work, we propose a methodology that enables the design and exploration of new loop scheduling strategies. In this methodology, a simulator is used to evaluate the most relevant existing scheduling strategies, and a genetic algorithm is employed to explore the solution space of the problem itself. The proposed methodology allowed us to design a new loop scheduling strategy, which showed to be up to 32.3x better than the existing policies in terms of load balance.
Complete list of metadatas

Cited literature [15 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01239916
Contributor : Jean-Francois Méhaut <>
Submitted on : Wednesday, November 15, 2017 - 7:18:10 PM
Last modification on : Thursday, February 7, 2019 - 2:27:58 PM
Long-term archiving on : Friday, February 16, 2018 - 3:53:08 PM

File

main.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01239916, version 1

Collections

Citation

Pedro Henrique Penna, Marcio Castro, Henrique Freitas, Francois Broquedis, Jean-François Méhaut. Uma Metodologia Baseada em Simulação e Algoritmo Genético para Exploração de Estratégias de Escalonamento de Laços. WSCAD 2015 - Simpósio em Sistemas Computacionais de Alto Desempenho, SBC, Oct 2015, Florianópolis, Brazil. ⟨hal-01239916⟩

Share

Metrics

Record views

352

Files downloads

37