Um Sistema de Comunicação para Multicomputadores

  • Gustavo Peixoto de Azevedo UFRJ

Abstract


This article presents a proposal of a communication system for large scale Multicomputers. The system’s main objective is to provide the communication between the tasks that make up an Parallel Distributed Application running on a multicomputer. The main contribuition of this work is the proposal of a new protocol that suports intertask communication between migrating tasks on a large scale multicomputer. This article presents a multicomputer communicating system’s fundamental properties, performance meters, basic structure, previous solutions and finally the solutions adoped in this work with special enfasis on the new protocol proposed for the communication between migrating tasks. The new protocol for intertask communication is defined, its correction is verified and its quality is checked against the communication system’s fundamental properties.

References

Azevedo G. P. de; "Escalonamento Dinâmico Distribuído em Multicomputadores de Alto Desempenho"; Dissertação de tese de Mestrado, Programa de Engenharia de Sistemas e Computação da COPPE/UFRJ, maio de 1992.

Dally W.; Network and processor architecture for message driven computers; VLSI and Parallel Computation, capítulo 3, 1989.

Douglis, F. e J. K. Ousterhout; Process Migration in the Sprite Operating System; Seventh International Conference on Distributed Computing, pp. 18-25, Sept. 1987.

Fowler, R. J.; Decentralized Object Finding Using Forwarding Addresses; PhD thesis, University of Washington, Seatle, Washington, December 1985.

Fowler, R. J.; The Complexity of Using Forwarding Addresses for Decentralized Object Finding; Proceedings of the Fifth ACM Symposium on the Principles of Distributed Computation, Calgary, Canada, August 1986.

Jul, E., H. Levy, N. Hutchinson e A. Black; Fine-Grain Mobility in the Emerald System; ACM Transactions on Computer Systems, Vol. 6, No. 1, February 1988.

Kermani P. e Kleinrock.L.; Virtual cut-through: a nnew computer communication switchig technnique; North-Holland, Computer Networks, Vol 3, No 4, pp. 267-286, September 1979.

Linder D. H e Harden J C.; An adaptative and fault tolerant wormhole routing strategy for k-ary n-cubes; IEEE Transactions on Computers, Vol 40, No 1, pp. 2-12, January 1991.

Lu C., Chen A. e Liu J.; Protocols for reliable process migration; In Proceedings of INFOCOM'87, San Francisco, California, March 1987.

Macharia G. M.; CLD: A novel approach to dynamic load balancing; Microprocessing and Microprogramming 28, pp. 43-48, 1989.

NiL. M., Xu C.-W. e Gendreau T. B.; A distributed drafting algorithm for load balancing; IEEE Transactions on Software Engineering SE-11(10), pp. 1153-1161, October 1985.

Ousterhout, J. K. et al.; The Sprite Network Operating System; Computer, February 1988.

Popek, G. J. e B. J. Walker; The LOCUS Distributed System Architecture; Computer Systems Serires, The MIT Press, 1985.

Powell, M. L. e B. P. Miller; Process Migration in DEMOS/MP; Proceedings of the Ninth Symposium on Operating Systems Principles, October 1983.

Ravi T. M. e Jefferson D.; A basic protocol for routing messages to migratin processes; In proceedings of the International Conference on Parallel Processeing, Vol. II, Software, August 1988.

Reed D. A. e Fujimoto R. M.; Multicomputer networks: message based parallel processing; MIT Press series in scientific computation, 1987.

Seitz C. et al; Wormhole chip project report; Inverno 1985.

Theimer, M.; Preemptable remote execution facilities for loosely-coupled distributed systems; Stanford University Technical Report STAN-CS-86-1128, June 1986.
Published
1993-09-07
AZEVEDO, Gustavo Peixoto de. Um Sistema de Comunicação para Multicomputadores. In: INTERNATIONAL SYMPOSIUM ON COMPUTER ARCHITECTURE AND HIGH PERFORMANCE COMPUTING (SBAC-PAD), 5. , 1993, Florianópolis/SC. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 1993 . p. 546-561. DOI: https://doi.org/10.5753/sbac-pad.1993.23058.