PLATINUM: A Placement Tool Based on Process Initiative
This paper presents a new load sharing method based on process initiative. The method can be applied either for individual pro- cesses in networks of workstations or for parallel computation. We know that the global computing power is most of the time much underutilized. Our goal is to share the load among the different processors. Usually, load sharing algorithms are based on the fact that information and de- cision of distribution are on the processor. We aim to give processes the abilityt o decide where they are going to run. This method is based on the independence and the initiative of entities. We have written behav- ior algorithms for independent processes. We have used them to realize a dynamic placement tool on a network of UNIX workstations : PLAT- INUM. The aim of this paper is to present our new method through PLATINUM.
KeywordsLoad Balance Load Share Communication Rate Distribute Computing System Load Indicator
Unable to display preview. Download preview PDF.
- 1.Arnold. O Allen: Probability, statistics and queuing theory with computer science applications. Academic Press. 1990.Google Scholar
- 2.Rajeev Motwani and Prabhakar Raghavan: Randomized Algorithms. Cambridge UniversityPress. 1995.Google Scholar
- 3.Andrea C. Arpaci-Dusseau and David E. Culler and Alan Mainwaring: Scheduling with Implicit Information in Distributed Systems. 1998 SIGMETRICS Conference on the Measurement and Modeling of Computer Systems, 233–243. June 24-26, 1998. Madison, Wisconsin.Google Scholar
- 4.D. J. Farber: The Distributed Computing System. Compcon Spring 73, pages 31–34. 1973.Google Scholar
- 5.Bertil Folliot: Méthodes et Outils de Partage de Charge pour la Conception et la Mise en Oeuvre d’Applications dans les Systémes Répartis Hétérogénes. Institute Blaise Pascal, MASI. 1993.Google Scholar
- 6.D. Ferrari and S. Zhou: An Empirical Investigation of Load Indices for Load Balancing Applications. Performances 87, 515–528. Bruxelles, Belgique.Google Scholar
- 7.Mor Harchol-Balter and Allen B. Downey: Exploiting Process Lifetime Distributions for Dynamic Load Balancing. ICSI. University of California at Berkeley, 1995.Google Scholar
- 8.M. J. Litzkow and M. Livny and M. W. Mutka: Condor: A Hunter of Idle Workstations. 8th International Conference on Distributed Computing Systems. IEEE Computer Society Press, 104–111. Washington, D. C., USA, 1988. 504Google Scholar
- 9.James S. Plank and Micah Beck and Gerry Kingsley: Libckpt: Transparent Checkpointing under UNIX. USENIX Winter 1995 Technical Conference. New Orleans Louisiana, 1995.Google Scholar