Análise de Desempenho Quantitativo do Sistema de Gerenciamento de Tarefas OpenPBS sobre um aglomerado de Pcs
Resumo
Este artigo apresenta a avaliação de desempenho quantitativa do sistema de gerenciamento de tarefas OpenPBS sobre uma máquina paralela baseada em aglomerados de PCs. A metodologia adotada para este fim é baseada em diferentes métodos descritos na literatura. Utilizando cargas de trabalho (“workloads”) dinâmicas analisa-se o desempenho do OpenPBS em função do tamanho das tarefas (com respeito ao numero de processadores que utilizam), do tempo de execução das tarefas e de diferente tipos de carga. Utilizando cargas de trabalho sintéticas mostra-se o overhead do OpenPBS e também ilustra-se o funcionamento do algoritmo de escalonamento do OpenPBS.Referências
Kento Aida, “Effect of Job Size Characteristics on Job Scheduling Performance”. In: Job Scheduling Strategies for Parallel Processing, Lecture Notes in Computer Science Vol. 1911, Springer-Verlag, pp.1-17, 2000.
Carsten Ernemann, Volker Hamscher, Uwe Schwiegelshohn, Ramin Yahyapour, Achim Streit, “On Advantages of Grid Computing for Parallel Job Scheduling”. In d
Eitan Frachtenberg, Dror G. Feitelson, Juan Fernandez, and Fabrizio Petrini, “Parallel Job Scheduling Under Dynamic workloads”. In Job Scheduling Strategies for Parallel Processing, Dror G. Feitelson, Larry Rudolph, and Uwe Schwiegelshohn, (ed.), Springer Verlag, Lect. Notes Comput. Sci. vol. 2862, pp. 208--227, 2003.
Eitan Frachtenberg, Fabrizio Petrini, Juan Fernandez, Scott Pakin, and Salvador Coll, “STORM: Lightning-Fast Resource Management”. In Supercomputing 2002.
Kris Gaj, Nguyen Nguyen, Jacek R. Radzikowski, Tarek El-Ghazawi, Nikitas Alexandridis, Frederic Vroman, Preeyapong Samipagdi, Suboh A. Suboh, “Performance Evaluation of Selected Job Management Systems”. In International Parallel and Distributed Processing Symposium: IPDPS 2002 Workshops
Rajkumar Kettimuthu, Vijay Subramani, Srividya Srinivasan, Thiagaraja Gopalasamy, “Selective Preemption Strategies for Parallel Job Scheduling”. In 2002 International Conference on Parallel Processing (ICPP'02)
Uri Lublin, Dror G. Feitelson, “The workload on parallel supercomputers: modeling the characteristics of rigid jobs”. In: Journal of Parallel and Distributed Computing,1105-1122 Volume 63, Number 1, January 2003
Jahanzeb Sherwani, Nosheen Ali, Nausheen Lotia, Zahra Hayat and Rajkumar Buyya, “Libra: An Economy driven Job Scheduling System for Aglomeradoss”. In HPC Asia 2002
Martha Torres, Alfredo Goldman, Junior Barrera, “A Parallel Algorithm for Enumerating Combinations”. In: 2003 International Conference on Parallel Processing.
Carsten Ernemann, Volker Hamscher, Uwe Schwiegelshohn, Ramin Yahyapour, Achim Streit, “On Advantages of Grid Computing for Parallel Job Scheduling”. In d
Eitan Frachtenberg, Dror G. Feitelson, Juan Fernandez, and Fabrizio Petrini, “Parallel Job Scheduling Under Dynamic workloads”. In Job Scheduling Strategies for Parallel Processing, Dror G. Feitelson, Larry Rudolph, and Uwe Schwiegelshohn, (ed.), Springer Verlag, Lect. Notes Comput. Sci. vol. 2862, pp. 208--227, 2003.
Eitan Frachtenberg, Fabrizio Petrini, Juan Fernandez, Scott Pakin, and Salvador Coll, “STORM: Lightning-Fast Resource Management”. In Supercomputing 2002.
Kris Gaj, Nguyen Nguyen, Jacek R. Radzikowski, Tarek El-Ghazawi, Nikitas Alexandridis, Frederic Vroman, Preeyapong Samipagdi, Suboh A. Suboh, “Performance Evaluation of Selected Job Management Systems”. In International Parallel and Distributed Processing Symposium: IPDPS 2002 Workshops
Rajkumar Kettimuthu, Vijay Subramani, Srividya Srinivasan, Thiagaraja Gopalasamy, “Selective Preemption Strategies for Parallel Job Scheduling”. In 2002 International Conference on Parallel Processing (ICPP'02)
Uri Lublin, Dror G. Feitelson, “The workload on parallel supercomputers: modeling the characteristics of rigid jobs”. In: Journal of Parallel and Distributed Computing,1105-1122 Volume 63, Number 1, January 2003
Jahanzeb Sherwani, Nosheen Ali, Nausheen Lotia, Zahra Hayat and Rajkumar Buyya, “Libra: An Economy driven Job Scheduling System for Aglomeradoss”. In HPC Asia 2002
Martha Torres, Alfredo Goldman, Junior Barrera, “A Parallel Algorithm for Enumerating Combinations”. In: 2003 International Conference on Parallel Processing.
Publicado
31/07/2004
Como Citar
TORRES, Martha; CHARRY, Edgar.
Análise de Desempenho Quantitativo do Sistema de Gerenciamento de Tarefas OpenPBS sobre um aglomerado de Pcs. In: WORKSHOP EM DESEMPENHO DE SISTEMAS COMPUTACIONAIS E DE COMUNICAÇÃO (WPERFORMANCE), 3. , 2004, Salvador/BA.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2004
.
p. 1-17.
ISSN 2595-6167.
