Load Balancing Algorithm Based On Task Transfer and Information Policy

Abstract

Cloud computing is a modern paradigm to provide services through the Internet. Nowadays, cloud computing is emerging field in information technology, next generation of computing. It provides very extensive measure of computing and storage Service gave to users through the internet which follows pay-as-you-go model. Major problems faced in the cloud are resource discovery, fault tolerance, load balancing and security. Load balancing is one of the main challenges, important technique, and critical issue and plays an important role which is required to distribute workload or task equally across the nodes or servers. Load balancing is a key aspect of cloud computing and avoids the situation in which some nodes become overloaded while the others are idle or have little work to do. Load balancing can improve the Quality of Service (QoS) metrics, including response time, cost, throughput, performance and resource utilization. Scheduling in Cloud computing infrastructure contain several challenging issues like computation time, budget, load balancing etc. Out of them, load balancing is one the major challenges for Cloud platform. Load balancing basically balances the load to achieve higher throughput and better resource utilization. Since scheduling task is NP-complete problem, so heuristic and meta heuristic approaches are preferred options to solve the same. In this dissertation, we chose a meta-heuristic approach to solve the task scheduling problem in cloud environment focussing mainly on two objectives, i.e., minimizing the makespan/ computation time and better load balancing