Ant Colony Optimization for Job Scheduling in Grid with Alea Simulator

Achieving high performance Grid scheduling in heterogeneous computing environment is critical. The Grid scheduling problem is an NP-complete problem. Because of its key importance on performance, the grid-scheduling problem in general has been extensively studied and various heuristics have been proposed. These heuristics are classified into a variety of categories such as job-Scheduling algorithms, Local-search algorithms, Duplication-based algorithms and Random based algorithms. Except for a few, these heuristics are mainly for system with fully connected homogeneous processors. The Ant Colony Algorithm has performed best compared to MET, MCT, OLB, MIN-MIN, MIN-MAX scheduling algorithms [8]. Problem with this algorithm is that it does not consider any resource failure and also does not consider CPU load at runtime. With the comparison of local search algorithms like First Come First Served (FCFS), (EDF) Earliest Deadline First, PBS (Priority Based Scheduling), EDF performs best in new extended gridsim toolkit called Alea[6]. We use an existing Ant Colony Optimization algorithm to perform job scheduling, in Alea Simulator and compare it with FCFS, EDF and PBS Scheduling algorithms. Citation: Chinmay Joshi, A D Patel Institute of Technology; Prerak Thakkar ,A D Patel Institute of Technology; Gopi Bhatt ,A D Patel Institute of Technology; Aniruddh Kurtkoti ,A D Patel Institute of Technology; Siddharth Shah ,A D Patel Institute of Technology. "Ant Colony Optimization for Job Scheduling in Grid with Alea Simulator." Global Research and Development Journal For Engineering 25 2017: 350 - 361.