Skip to Main Content
The Grid can be described as a distributed computing infrastructure for coordinated resource sharing and problem solving in dynamic, multi-institutional virtual organizations. Since most of the applications in grid computing field fall into interdependent task model called workflow applications and in order to utilize the available resources in grids efficiently and effectively, a lot of research have been made in resource management, especially in scheduling. The main objective of scheduling is the attenuation of application completion time with usage of redundant resources. In this paper, we propose a novel semi-dynamic scheduling algorithm called adaptive dual-objective scheduling (ADOS) algorithm which statically generates the initial schedule using an evolutionary technique and adapts it dynamically as the performance of resources changes. Scheduling is performed by the algorithm considering both completion time and resource usage with changes in performance of grid resources. We also evaluate the algorithm through set of experiments demonstrating that the proposed algorithm delivers promising performance in three respects: completion time, resource utilization, and robustness to resource-performance fluctuations. The performance of the proposed algorithm is measured by implementing it using a Grid simulator (Gridsim).