Task scheduling with load balancing using multiple ant colonies optimization in grid computing

Abstract

Task scheduling with load balancing in grid computing aims to assign tasks to computing nodes and minimize the execution time of tasks as well as workload across all nodes. Despite of the intractability, the scheduling problem is of particular concern to both users and grid systems. In this paper, a multiple ant colonies optimization (MACO) approach is proposed for achieving task scheduling with load balancing. In the MACO approach, multiple ant colonies work together and exchange information to collectively find solutions with a two-fold objective of minimizing the execution time of tasks and the degree of imbalance of computing nodes. Experimental results show that our algorithm outperforms FCFS and ACS approaches.