Applying a cost effective hybrid approach to task scheduling and matching for computational grids

To utilize computational grids more efficaciously, directed acyclic graph (DAG)-based approach is required. This paper falls into the context of applying cost effective approach to DAG scheduling to lower down the overall program completion time. The proposed algorithm, namely, Cost-effective Hybrid approach (CEHA), coalesces the foredeals of cuckoo search and ant colony optimization. Experimental analysis depicted the expeditious converging property of cuckoo search algorithm exploiting it to find sub-optimal solution which drives ACO to more optimized solution. CEHA finds the order of all tasks to be scheduled in a parallel program and then allocates these tasks to the most appropriate processing elements in a computational grid finally minimizing the total completion time of the parallel program. Simulations demonstrated that the proposed CEHA outperforms sundry algorithms like N. Moganarangan et al. method, Ant colony optimization algorithm, and Cuckoo search algorithm.