MODIFIED OPTIMAL ALGORITHM FOR LOAD BALANCING IN CLOUD COMPUTING

A cloud is a technology which provides services over the network on demand. In cloud computing, the cloud user may get access to the data and applications from anywhere at any time. Users have the advantage of paying only for what they are using satisfying their fluctuating demands. Managing such a large amount of user requests coming from all over the globe is a tedious task, thus the cloud service provider should take efficient steps to handle them. This involves the concept of load balancing. Load Balancing basically helps in even distribution of loads among all the nodes such that no node is overloaded or under loaded. Considering the growing users and importance of cloud, finding new ways which can prove efficient in balancing the load among various processing nodes will always be an area of concern and research. Various algorithms already exist for load balancing, each having its own concept but have the same goal to reduce the response time to client problems. The goal of this paper is to propose an algorithm for Load balancing. In the existing optimized scheduling algorithm, Shortest Job First (SJF) should be used to arrange the tasks and then Round Robin (RR) should be used for processing, without giving any preference to their priorities preventing important tasks from getting the amount of attention they need. The algorithm proposed in this paper aims to give higher importance to requests with higher priorities.