A loosely coupled hybrid meta-heuristic algorithm for the static independent task scheduling problem in grid computing
Task scheduling is one of the most difficult problems in grid computing systems. Therefore, various studies have been proposed to present methods which provide efficient schedules. Meta-heuristic approaches are among the methods which have proven their efficiency in this domain. However, the literature shows that hybridizing two or more meta-heuristics can improve performance to a greater extent than stand-alone algorithms as the new high-level algorithm will inherit the best features of the hybridized algorithms. In this paper, a loosely coupled hybrid meta-heuristic algorithm is proposed for solving the static independent task scheduling problem in grid computing. It combines ant colony optimization and variable neighborhood search, where the former operates first and whose output is subsequently improved by the latter. The experimental results show that the proposed algorithm achieves better task-machine mapping in terms of minimizing makespan than other selected approaches from the literature.
Citation : Younis, M. T, Yang, S. and Passow, B. N. (2018), 'A loosely coupled hybrid meta-heuristic algorithm for the static independent task scheduling problem in grid computing'. 2018 IEEE Congress on Evolutionary Computation (CEC). Barra da Tijuca, Rio de Janeiro, Brazil , 8-13 July.
Research Group : Centre for Computational Intelligence
Research Institute : Institute of Artificial Intelligence (IAI)
Peer Reviewed : Yes