Memory Constrained ANT Colony System for Task Scheduling in GRID Computing

Grid computing solves the increasing need of scientific, engineering and research problems. It combines the geographically distributed resources to solve a computation intensive problem which cannot be solved using a single resource. Resource sharing requires more optimized algorithmic structure, otherwise the response time is increased and the resource utilization is reduced. In order to avoid such reduction in the performance of the grid system, an optimal resource sharing algorithm is required. The ACO solves many engineering problems and provides optimal result which includes Travelling Salesman Problem, Network Routing, and Scheduling. This paper proposes Load Shared Ant Colony Optimization (LSACO) which shares the load among the available resources. The proposed method considers memory requirement as a parameter to distribute the load among Grid resources. LSACO reduces the overall response time and increases the resource utilization and number of tasks scheduled. The proposed method has been tested for different types of tasks and resources.