Análise de Características Comportamentais de Aplicações OpenMP para Redução do Consumo de Energia
Resumo
Desempenho e consumo energético são requisitos fundamentais em sistemas de computação. Um desafio comumente encontrado é conciliar esses dois aspectos, buscando manter o mesmo desempenho, consumindo cada vez menos energia. Muitas técnicas possibilitam a redução do consumo de energia em aplicações paralelas, mas na maioria das vezes elas envolvem recursos encontrados apenas em processadores modernos ou um conhecimento amplo das características da aplicação e da plataforma alvo. Nesse trabalho propomos uma biblioteca que altera dinamicamente a frequência de processador, utilizando como base de conhecimento uma análise prévia das regiões limitadas pela memória. Os resultados apresentam uma redução de 1,89% no consumo de energia para aplicação Lulesh com cerca de 0,09% de sobrecarga no tempo de execução, quando comparamos nossa técnica com a política de processador Ondemand do Sistema Operacional Linux.
Referências
Brodowski, D., Golde, N., Wysocki, R. J., and Kumar, V. (2013). Linux cpufreq governors – information for users and developers. Technical report, Linux.
Diamond, J., Burtscher, M., McCalpin, J. D., Kim, B.-D., Keckler, S. W., and Browne, J. C. (2011). Evaluation and optimization of multicore performance bottlenecks in supercomputing applications. In Performance Analysis of Systems and Software (ISPASS), 2011 IEEE International Symposium on, pages 32–43. IEEE.
Hutcheson, A. and Natoli, V. (2011). Memory bound vs. compute bound: A quantitative study of cache and memory bandwidth in high performance applications. Technical report, Stone Ridge Technology.
Jesshope, C. and Egan, C. (2006). Advances in Computer Systems Architecture: 11th Asia-Pacific Conference, ACSAC 2006, Shanghai, China, September 6-8, 2006, Proceedings. LNCS sublibrary: Theoretical computer science and general issues. Springer.
Karlin, I., Keasler, J., and Neely, R. (2013). Lulesh 2.0 updates and changes. Technical Report LLNL-TR-641973, LLNL.
Laurenzano, M. A., Meswani, M., Carrington, L., Snavely, A., Tikir, M. M., and Poole, S. (2011). Reducing energy usage with memory and computation-aware dynamic frequency scaling. In European Conference on Parallel Processing, pages 79–90. Springer.
LLNL (2018). Hydrodynamics Challenge Problem, Lawrence Livermore National Laboratory. Technical Report LLNL-TR-490254, LLNL.
Millani, L. F. and Schnorr, L. M. (2016). Computation-aware dynamic frequency scaling: Parsimonious evaluation of the time-energy trade-off using design of experiments. In European Conference on Parallel Processing, pages 583–595. Springer.
Molka, D., Schöne, R., Hackenberg, D., and Nagel, W. E. (2017). Detecting memory-boundedness with hardware performance counters. In Proceedings of the 8th ACM/SPEC on International Conference on Performance Engineering, pages 27–38. ACM.
Moro, G. B. and Schnorr, L. M. (2017). Detecting memory-bound parallel regions to improve the energy-efficiency of applications. In XIV Workshop de Processamento Paralelo e Distribuído, pages 10–13.
Orgerie, A.-C., de Assuncao, M. D., and Lefevre, L. (2014). A Survey on Techniques for Improving the Energy Efficiency of Large-scale Distributed Systems. ACM Comput. Surv., 46(4):47:1–47:31.
Shafik, R. A., Das, A., Yang, S., Merrett, G., and Al-Hashimi, B. M. (2015). Adaptive energy minimization of openmp parallel applications on many-core systems. In Proceedings of the 6th Workshop on Parallel Programming and Run-Time Management Techniques for Many-core Architectures, pages 19–24. ACM.
Silveira, D. S., Moro, G. B., Cruz, E. H. M., Navaux, P. O. A., Schnorr, L. M., and Bampi, S. (2016). Energy Consumption Estimation in Parallel Applications: an Analysis in Real and Theoretical Models. In WSCAD 2016 - XVII Simpósio em Sistemas Computacionais de Alto Desempenho, pages 134–145.
Subramanian, L. (2015). Providing high and controllable performance in multicore systems through shared resource management. arXiv preprint arXiv:1508.03087.
Sueur, E. L. and Heiser, G. (2010). Dynamic voltage and frequency scaling: the laws of diminishing returns. In Proceedings of the 2010 international conference on Power aware computing and systems, pages 1–8.
Wang, W., Porterfield, A., Cavazos, J., and Bhalachandra, S. (2015). Using per-loop cpu clock modulation for energy efficiency in openmp applications. In Parallel Processing (ICPP), 2015 44th International Conference on, pages 629–638. IEEE.
Diamond, J., Burtscher, M., McCalpin, J. D., Kim, B.-D., Keckler, S. W., and Browne, J. C. (2011). Evaluation and optimization of multicore performance bottlenecks in supercomputing applications. In Performance Analysis of Systems and Software (ISPASS), 2011 IEEE International Symposium on, pages 32–43. IEEE.
Hutcheson, A. and Natoli, V. (2011). Memory bound vs. compute bound: A quantitative study of cache and memory bandwidth in high performance applications. Technical report, Stone Ridge Technology.
Jesshope, C. and Egan, C. (2006). Advances in Computer Systems Architecture: 11th Asia-Pacific Conference, ACSAC 2006, Shanghai, China, September 6-8, 2006, Proceedings. LNCS sublibrary: Theoretical computer science and general issues. Springer.
Karlin, I., Keasler, J., and Neely, R. (2013). Lulesh 2.0 updates and changes. Technical Report LLNL-TR-641973, LLNL.
Laurenzano, M. A., Meswani, M., Carrington, L., Snavely, A., Tikir, M. M., and Poole, S. (2011). Reducing energy usage with memory and computation-aware dynamic frequency scaling. In European Conference on Parallel Processing, pages 79–90. Springer.
LLNL (2018). Hydrodynamics Challenge Problem, Lawrence Livermore National Laboratory. Technical Report LLNL-TR-490254, LLNL.
Millani, L. F. and Schnorr, L. M. (2016). Computation-aware dynamic frequency scaling: Parsimonious evaluation of the time-energy trade-off using design of experiments. In European Conference on Parallel Processing, pages 583–595. Springer.
Molka, D., Schöne, R., Hackenberg, D., and Nagel, W. E. (2017). Detecting memory-boundedness with hardware performance counters. In Proceedings of the 8th ACM/SPEC on International Conference on Performance Engineering, pages 27–38. ACM.
Moro, G. B. and Schnorr, L. M. (2017). Detecting memory-bound parallel regions to improve the energy-efficiency of applications. In XIV Workshop de Processamento Paralelo e Distribuído, pages 10–13.
Orgerie, A.-C., de Assuncao, M. D., and Lefevre, L. (2014). A Survey on Techniques for Improving the Energy Efficiency of Large-scale Distributed Systems. ACM Comput. Surv., 46(4):47:1–47:31.
Shafik, R. A., Das, A., Yang, S., Merrett, G., and Al-Hashimi, B. M. (2015). Adaptive energy minimization of openmp parallel applications on many-core systems. In Proceedings of the 6th Workshop on Parallel Programming and Run-Time Management Techniques for Many-core Architectures, pages 19–24. ACM.
Silveira, D. S., Moro, G. B., Cruz, E. H. M., Navaux, P. O. A., Schnorr, L. M., and Bampi, S. (2016). Energy Consumption Estimation in Parallel Applications: an Analysis in Real and Theoretical Models. In WSCAD 2016 - XVII Simpósio em Sistemas Computacionais de Alto Desempenho, pages 134–145.
Subramanian, L. (2015). Providing high and controllable performance in multicore systems through shared resource management. arXiv preprint arXiv:1508.03087.
Sueur, E. L. and Heiser, G. (2010). Dynamic voltage and frequency scaling: the laws of diminishing returns. In Proceedings of the 2010 international conference on Power aware computing and systems, pages 1–8.
Wang, W., Porterfield, A., Cavazos, J., and Bhalachandra, S. (2015). Using per-loop cpu clock modulation for energy efficiency in openmp applications. In Parallel Processing (ICPP), 2015 44th International Conference on, pages 629–638. IEEE.
Publicado
22/07/2018
Como Citar
MORO, Gabriel B.; SCHNORR, Lucas Mello.
Análise de Características Comportamentais de Aplicações OpenMP para Redução do Consumo de Energia. In: WORKSHOP EM DESEMPENHO DE SISTEMAS COMPUTACIONAIS E DE COMUNICAÇÃO (WPERFORMANCE), 17. , 2018, Natal.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2018
.
p. 99-111.
ISSN 2595-6167.
DOI: https://doi.org/10.5753/wperformance.2018.3346.
