Survey and Analysis of Scheduling Algorithms for Dynamic Load Balancing in Computational Grids, Cloud and Cluster

A computational grid is a large scale, heterogeneous collection of autonomous systems, geographically distributed and interconnected by heterogeneous networks [6]. The user’s resource requirements in the grids vary depending on their goals, time constraints, priorities and budgets [3]. A cluster has three basic elements�a collection of individual computers, a network connecting those computers, and software that enables a computer to share work among the other computers via the network. Cloud computing is a new technology .It providing online resources and online storage to the user’s. In cloud computing users can access resources all the time through internet. They need to pay only for those resources as much they use. Partitioning and load-balancing algorithms compute the distribution of application data and work to processors. The load balancing goal is to fully utilize the computing power from multiple hosts without the disturbing the user, regardless of the number of hosts available in the background and aims to improve the overall performance [4]. Load balancing helps to distribute all loads between all the nodes. It also ensures that every computing resource is distributed efficiently and fairly. Scheduling is a main concept in computer multitasking and multiprocessing operating system. Scheduling refers to the way in which processes are assigned to run on the available CPUs. There are various scheduling algorithms in computational grid, cloud and cluster.