Genetic Algorithm for Process Planning in Distributed Systems with Arbitrary Topology

maxympinchuk.jpgMaxym Pinchuk is a graduate of the Kharkiv National University of Radioelectronics (KNURE), Ukraine and has recently passed his MPhil viva at Wessex Institute of Technology with a thesis entitled “Genetic algorithm for process planning in distributed systems with arbitrary topology”. The external examiner was Dr Daniel Kudenko from the University of York, UK and the internal examiner was Prof Alexander Galybin.

The scope of Maxym’s thesis is in the field of Evolution computation applied to process mapping in distributed systems. The main goals were to analyze the possibility of mapping tasks by means of a genetic algorithm, to analyze the dependency of convergence of the genetic algorithm from its parameters on different tasks, and to recommend the most appropriate operators and parameters of the algorithm.

The mapping algorithm is one of the fundamental elements of managing and load planning of each cluster system. The mapping algorithm must trade load balancing for the reduction of communication, and vice versa.

As a result of this investigation, a heuristic approach of mapping tasks onto distributed systems was developed. The correctness and effectiveness of the algorithm was proved on the base of comparison with the other techniques.

Maxym was congratulated by the examiners on the success of his research and recommended the award of Master of Philosophy.

Grateful acknowledgement is given to the Foreign and Commonwealth Office for their support.