Aprendizado por Reforço aplicado a escalonamento em Grids
Resumo
Aprendizado por reforço é uma técnica simples que possui aplicação em várias áreas. Um ambiente real de grid, em geral dinâmico e heterogêneo, oferece um ambiente interessante para sua aplicação. Neste trabalho, utilizamos esta técnica para classificar os nós disponíveis em um grid, dando suporte assim a dois algoritmos de escalonamento, AG e MQD. Um ambiente de grid real foi montado e experimentos foram realizados com estes dois algoritmos, de maneira a verificar seu impacto em um ambiente real, com e sem a presença de reescalonamento.Referências
R. Buyya, D. Abramson, and J. Giddy. Nimrod/g: An architecture for a resource management and scheduling system in a global computational grid. hpc, 01:283, 2000.
H. Casanova. Simgrid: A toolkit for the simulation of application scheduling. In CCGRID, page 430. IEEE, 2001.
F. Dong and S. Akl. Scheduling algorithms for grid computing: State of the art and open problems. Technical Report 2006-504, School of Computing, Queen’s University, Jan 2006.
R. M. E. Huedo and I. Llorente. The gridway framework for adaptive scheduling and execution on grids. Scalable Computing - Practice and Experience, 6(3):1–8, Sep 2005.
C. B. et al. An easygrid portal for scheduling system-aware applications on computational grids. Concurrency and Computation: Practice and Experience, 18(6):553–566, 2006.
D. P. S. et al. Trading cycles for information: Using replication to schedule bag-of-tasks applications on computational grids. In Euro-Par, pages 169–180, 2003.
F. B. et al. New grid scheduling and rescheduling methods in the grads project. Int. J. Parallel Program., 33(2):209–229, 2005.
H. C. et al. The apples parameter sweep template: user-level middleware for the grid. In Supercomputing, page 60. IEEE, 2000.
P. K. V. et al. Grand: Toward scalability in a grid environment. Concurrency and Computation: Practice and Experience, 19(14):1991–2009, 2007.
A. Galstyan, K. Czajkowski, and K. Lerman. Resource allocation in the grid using reinforcement learning. In AAMAS, pages 1314–1315. IEEE, 2004.
Y. C. Lee and A. Y. Zomaya. A grid scheduling algorithm for bag-of-tasks applications using multiple queues with duplication. icis-comsar, 0:5–10, 2006.
K. Nadiminti, S. Venugopal, H. Gibbins, T. Ma, and R. Buyya. The gridbus grid service broker and scheduler (2.4.4) user guide. http://www.gridbus.org/broker/2.4.4/manualv2.4.4.pdf, Agosto 2007.
R. Sutton and A. Barto. Reinforcement Learning: An Introduction. MIT Press, Cambridge, MA, 1998.
S. Venugopal, R. Buyya, and L. Winton. A grid service broker for scheduling distributed data-oriented applications on global grids. In MGC. ACM Press, 2004.
C. J. C. H. Watkins and P. Dayan. Technical note: Qlearning. Mach. Learn., 8(3-4):279–292, 1992.
H. Casanova. Simgrid: A toolkit for the simulation of application scheduling. In CCGRID, page 430. IEEE, 2001.
F. Dong and S. Akl. Scheduling algorithms for grid computing: State of the art and open problems. Technical Report 2006-504, School of Computing, Queen’s University, Jan 2006.
R. M. E. Huedo and I. Llorente. The gridway framework for adaptive scheduling and execution on grids. Scalable Computing - Practice and Experience, 6(3):1–8, Sep 2005.
C. B. et al. An easygrid portal for scheduling system-aware applications on computational grids. Concurrency and Computation: Practice and Experience, 18(6):553–566, 2006.
D. P. S. et al. Trading cycles for information: Using replication to schedule bag-of-tasks applications on computational grids. In Euro-Par, pages 169–180, 2003.
F. B. et al. New grid scheduling and rescheduling methods in the grads project. Int. J. Parallel Program., 33(2):209–229, 2005.
H. C. et al. The apples parameter sweep template: user-level middleware for the grid. In Supercomputing, page 60. IEEE, 2000.
P. K. V. et al. Grand: Toward scalability in a grid environment. Concurrency and Computation: Practice and Experience, 19(14):1991–2009, 2007.
A. Galstyan, K. Czajkowski, and K. Lerman. Resource allocation in the grid using reinforcement learning. In AAMAS, pages 1314–1315. IEEE, 2004.
Y. C. Lee and A. Y. Zomaya. A grid scheduling algorithm for bag-of-tasks applications using multiple queues with duplication. icis-comsar, 0:5–10, 2006.
K. Nadiminti, S. Venugopal, H. Gibbins, T. Ma, and R. Buyya. The gridbus grid service broker and scheduler (2.4.4) user guide. http://www.gridbus.org/broker/2.4.4/manualv2.4.4.pdf, Agosto 2007.
R. Sutton and A. Barto. Reinforcement Learning: An Introduction. MIT Press, Cambridge, MA, 1998.
S. Venugopal, R. Buyya, and L. Winton. A grid service broker for scheduling distributed data-oriented applications on global grids. In MGC. ACM Press, 2004.
C. J. C. H. Watkins and P. Dayan. Technical note: Qlearning. Mach. Learn., 8(3-4):279–292, 1992.
Publicado
29/10/2008
Como Citar
COSTA, Bernardo Fortunato; DUTRA, Inês; MATTOSO, Marta.
Aprendizado por Reforço aplicado a escalonamento em Grids. In: SIMPÓSIO EM SISTEMAS COMPUTACIONAIS DE ALTO DESEMPENHO (SSCAD), 9. , 2008, Campo Grande.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2008
.
p. 109-116.
DOI: https://doi.org/10.5753/wscad.2008.17674.
