Load Balancing with a Fuzzy-Decision Algorithm

The University of Dortmund is a member of the LYDIA project and deals within the project with the possible use of decision algorithms for load balancing. In a simple case, one can grasp each kind of distribution as a decision problem, at which it must be decided which job should be distributed to which node in the system. This idea will be shortly discussed in this article, but we will concentrate on a further developed approach. Within this approach, there is a situation-specific selection of load balancing algorithms. The main idea of the approach is outlined in this article and a suitable decision algorithm will be presented. In addition to that, some results of simulations will be shown and there will be also an outlook to further activities.

[1]  Rudolf Felix Goal-oriented control of VLSI design processes based on fuzzy sets , 1990, Proceedings of the Twentieth International Symposium on Multiple-Valued Logic.

[2]  Karl-Heinz Temme,et al.  Hybrid Reasoning with FUZZYNEX , 1994 .

[3]  H. Zimmermann,et al.  Fuzzy Set Theory and Its Applications , 1993 .

[4]  S. L. Hurst,et al.  9th International Symposium on Multiple-Valued Logic , 1978 .

[5]  Rudolf Kruse,et al.  Fuzzy-systems in computer science , 1994 .

[6]  Donald F. Ferguson,et al.  Satisfying response time goals in transaction processing systems , 1993 .

[7]  Lotfi A. Zadeh,et al.  Fuzzy Sets , 1996, Inf. Control..

[8]  C. Sierra,et al.  A new approach to connective generation in the framework of expert systems using fuzzy logic , 1988, [1988] Proceedings. The Eighteenth International Symposium on Multiple-Valued Logic.

[9]  E. H. Mamdani,et al.  Advances in the linguistic synthesis of fuzzy controllers , 1976 .

[10]  John A. Stankovic,et al.  An Application of Bayesian Decision Theory to Decentralized Control of Job Scheduling , 1985, IEEE Transactions on Computers.

[11]  A. Hoffmann,et al.  VLSI chip-architecture selection using reasoning based on fuzzy logic , 1989, Proceedings. The Nineteenth International Symposium on Multiple-Valued Logic.