A GENETIC TASK ALLOCATION ALGORITHM FOR DISTRIBUTED COMPUTING SYSTEMSINCORPORATING PROBLEM SPECIFIC KNOWLEDGE

Distributed Computing Systems (DCS) promise a convenient platform for parallel processing and consequently can be expected to provide highly improved throughput and turnaround characteristics for all types of c omputing jobs. Task allocation in DCS remains to be an important and r elevant problem attracting the attention of researchers in the discipl ine. Genetic Algorithms (GA) have successfully been used to solve vari ous optimization problems. A GA based task allocation model for multip rocessors has been proposed by Hou, Ansari & Ren [3]. We present a Gen etic Task Allocation Algorithm for DCS, wherein we have considered the underlying interconnection network of the processors, communication r equirements among modules of the tasks apart from the precedence relat ion of the task graph that has been considered in [3] also. We have al so considered multiprogramming at every processing nodes with related characteristic values. We have, intentionally, made use of the finding [4] that the incorporation of the problem specific knowledge in const ruction of GAs improves the initial population structures. The model a nd algorithm proposed by us is sufficiently simple and adequately usab le for the purpose of simulation experiments and its possible incorpor ation in future operating systems of DCS.