Applying ant colony optimization for load balancing on grid

Abstract

Grid computing is comprised of many distributed nodes in order to compute and analyze a set of distributed data. To improve the processing performance, an appropriate load-balancing algorithm is required to equally distribute loads among the grids nodes. In this article, an algorithm based on ant colony optimization is proposed to deal with load-balancing problems. In this approach, when an ant reaches a node, the ants table and the nodes table exchange their information and update each other. In order to move to the most appropriate node, the ant selects the next node from the current nodes table according to the nodes loads and their CPU rates. This process is continued until the ant passes the predefined steps. The experimental results show that while implementing the proposed algorithm to the grid environment, increasing the number of jobs and their length has insignificant impact on the system response time. © 2015 The Chinese Institute of Engineers.