Darwin: uma framework de otimização clusterizável
Resumo
Neste artigo apresentamos Darwin, uma ferramenta de otimização de aplicações a partir do uso de algoritmos meta-heurísticos. A ferramenta dispõe de três otimizações: algoritmo genético, particle swarm e evolução diferencial. Conta também com dois backends para execução, o primeiro focado no uso de um cluster e o segundo no uso de recursos computacionais locais. Utilizou-se o simulador de arquitetura x86 Sniper com o benchmark Parsec, otimizando-se parâmetros de cache, para validar a ferramenta desenvolvida. Ao final, avalia-se que a ferramenta permite a otimização utilizando somente algoritmos de forma simplificada e paralelizável.
Palavras-chave:
meta-heurística, cluster, otimização
Referências
Bienia, C. (2011). Benchmarking Modern Multiprocessors. PhD thesis, Princeton University.
Carlson, T. E., Heirman, W., Eyerman, S., Hur, I., and Eeckhout, L. (2014). An evaluation of high-level mechanistic core models. ACM Transactions on Architecture and Code Optimization (TACO).
Eberhart, R. C., Shi, Y., and Kennedy, J. (2001). Swarm intelligence. Elsevier.
Goldberg, D. E. (1989).Genetic Algorithms in Search, Optimization and Machine Learning. Addison-Wesley Longman Publishing Co., Inc., Boston, MA, USA, 1st edition.
Li, S., Ahn, J. H., Strong, R. D., Brockman, J. B., Tullsen, D. M., and Jouppi, N. P. (2009). Mcpat: an integrated power, area, and timing modeling framework for multicore and manycore architectures. In Proceedings of the 42nd Annual IEEE/ACM International Symposium on Microarchitecture, pages 469–480. ACM.
Papa, J. P., Rosa, G. H., Rodrigues, D., and Yang, X. (2017). Libopt: An open-source platform for fast prototyping soft optimization techniques. CoRR, abs/1704.05174.
Rossell, L. B. I. (2004–2019). ts - task spooler. a simple unix batch system. https://launchpad.net/ubuntu/+source/task-spooler/0.7.5-1.
Storn, R. and Price, K. (1997). Differential evolution–a simple and efficient heuristic for global optimization over continuous spaces. Journal of global optimization, 11(4):341–359.
Thain, D., Tannenbaum, T., and Livny, M. (2005). Distributed computing in practice: the condor experience. Concurrency and computation: practice and experience, 17(2-4):323–356.
Carlson, T. E., Heirman, W., Eyerman, S., Hur, I., and Eeckhout, L. (2014). An evaluation of high-level mechanistic core models. ACM Transactions on Architecture and Code Optimization (TACO).
Eberhart, R. C., Shi, Y., and Kennedy, J. (2001). Swarm intelligence. Elsevier.
Goldberg, D. E. (1989).Genetic Algorithms in Search, Optimization and Machine Learning. Addison-Wesley Longman Publishing Co., Inc., Boston, MA, USA, 1st edition.
Li, S., Ahn, J. H., Strong, R. D., Brockman, J. B., Tullsen, D. M., and Jouppi, N. P. (2009). Mcpat: an integrated power, area, and timing modeling framework for multicore and manycore architectures. In Proceedings of the 42nd Annual IEEE/ACM International Symposium on Microarchitecture, pages 469–480. ACM.
Papa, J. P., Rosa, G. H., Rodrigues, D., and Yang, X. (2017). Libopt: An open-source platform for fast prototyping soft optimization techniques. CoRR, abs/1704.05174.
Rossell, L. B. I. (2004–2019). ts - task spooler. a simple unix batch system. https://launchpad.net/ubuntu/+source/task-spooler/0.7.5-1.
Storn, R. and Price, K. (1997). Differential evolution–a simple and efficient heuristic for global optimization over continuous spaces. Journal of global optimization, 11(4):341–359.
Thain, D., Tannenbaum, T., and Livny, M. (2005). Distributed computing in practice: the condor experience. Concurrency and computation: practice and experience, 17(2-4):323–356.
Publicado
19/08/2020
Como Citar
RUSA, Henrique A.; KRUGER, Kleber; AZEVEDO, Rodolfo.
Darwin: uma framework de otimização clusterizável. In: ESCOLA REGIONAL DE ALTO DESEMPENHO DE SÃO PAULO (ERAD-SP), 11. , 2020, Evento Online.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2020
.
p. 106-109.
DOI: https://doi.org/10.5753/eradsp.2020.16898.
