An Overview of the Scheduling Policies and Algorithms in Grid Computing
P. Muthulakshmi · 2011
The advancement of computers and network technologies have taken us to the world of high per formance computing called the Grid computing which is the evolution of parallel processing. Rather the improvement in per formances is still expected. The orientation of grid computing i s to integrate and aggregate the resources contributing themselves in a geographically distributed environment. The view on such environments remains incomplete if we fail to look into scheduling. Scheduling is the ultimate goal aiming process of mapping the jobs submitted to the grid and the appr opriate resources available. This paper presents a package of reviews taking various factors that are having greater infl uence while scheduling the jobs. Here we consider the algorithm s keyed with communication cost, execution time length, error fa ctors, task duplications, data intensive, and heterogenic netwo rk behavior and so on. A sequence of classification and comparison of algorithms evolved from homogeneous scheduling to the state of art scheduling is listed. Though the earlier algorithms are the pioneers of scheduling, we have studied the perspectives, wo rking principles, the grid domain in which the algorithms have been applied. Key Terms: scheduling algorithms, scheduling policies, hybrid algorithms, dynamicity, heterogenic network behavior, grid computing, parallel processing