Adaptive job Scheduling for Computational Grid based on Ant Colony Optimization with Genetic Parameter Selection

Demand of new generation of internet technology applied a distributed process of computing for service providing and resource allocation. The service and resource allocation needed a computational grid for task processing. Computational grid manages a process of resource and task allocation process. The alloca tion of resource and task effect the performance of grid mechanism. For the scheduling of task and resource for computational grid used a queuing process model such as first come first served process. But the performance of this model is very impartial, no w various authors and researchers used a process of efficient searching technique for job allocation such as heuristic and meta - heuristic improved the performance of computational grid. But the uncontrolled nature of meta - heuristic such as ant colony optim ization degraded the performance of grid allocation. In this paper we proposed a controlled mechanism of ant colony optimization with genetic algorithm for task scheduling in computational grid. For the performance evaluation of computational grid we used 6 *6, 10*10 and 20 *5 grid parameter and the measurement of performance in terms of job completion and failure of job. Our empirical evaluation shows that better performance instead of ANT and PSO scheduling technique.