Energy-Aware Load Balanced Scheduler for Cloud Computing to Improve Quality of Service

Cloud computing paradigm has progressed as extremely ascendable services with increasing power on computation, massive storage possibility and the resources being obtained as service using cloud environment by providing guarantee to the Service Level. However, the subscriber requirement has increased to an extent that necessitates a big active platform for balancing the load though the available resources are shared according to the availability. With the advent of powerful network processors, the cloud computing paradigm also give rise to load balancing problem which has to be solved in an optimal manner in order to avoid deadlock and enhance resource effectiveness. To address the issue of energy conservation in cloud computing paradigm, the use of progressive traffic data from data centers applied a service invocation forecast model. The collaborative data possession scheme applied responsive pattern using the Homomorphic verifiable format and hierarchy was performed using hash index model but the disadvantage of the model was that the match index structure did not matched properly using cluster. However enforcing policies in cloud computing paradigm is challenging because of the different slabs of power tariffs and requirements made to the servers affect the decisions, whether the loads to be pushed in or out of a cluster affecting the overall energy utilization. To address the issue of balancing the load and optimizing the bandwidth and energy utilization, an Energy-aware Load Balanced Scheduler (ELBS) for cloud computing to improve Quality of Service is presented.. The effectiveness of the proposed ELBS model is illustrated by theoretical analysis with Virtual Machine (VM) energy-efficient cloud data centers. Performance metric for evaluation of ELBS model is measured in terms of computational complex for energy utilization, performance (in terms of throughput), bandwidth utilization rate, cloud computation cost, and response time to service invocation, load balance to improve QoS and clustering quality.