International Journal of Engineering and Technology

Cloud computing is the outcome of rapid growth of internet. Its future depends upon the optimum resource utilization, higher customer satisfaction and efficient installation of infrastructure. Due to ubiquitous, pay per usage, elasticity and lower installation cost many service providers are providing services through cloud. This has inundated the task at cloud service provider end. So, need of efficient and cloud compatible load balancing strategy has been felt. Load balancing is the major research area among the key challenges in the field of cloud computing. It involves mapping between task & resource in such a way that all resources are utilized in optimal manner. At the same time it is also desirable that load balancing approach must be scalable, fault tolerant and fault recoverable so that trust of cloud user will increase in the services provided by cloud service provider. Keeping all the desirable characteristics of cloud environment in consideration, authors have proposed weighted random walk based load balancing approach which not only increases the user trust in cloud services but also matches with the cloud essential characteristics. Weighted random walk based approach has implemented the biased random walk concept. Biasing has been implemented using weight of virtual machine which is based on virtual machine capacity and no of tasks in its queue.