Green Cloud: An Energy Efficient Load Balancing Approach Using Global Load Optimization

High Performance Green Cloud Computing is an emerging field in the upcoming IT industry. With the main focus on balancing resource utilization in green cloud environment, IT applications need a highly rapid and scalable access services. Cyber Guarder Approach (CGA) based on trust management in green cloud facilitates access control to resources pool but the energy expenditure is of higher rate while balancing the load. One of the important aspects of green cloud being optimization of energy utilization, Energy-aware Allocation Policies (EAP) maximized the revenues. However, EAP policies increased the running server, resulting in higher amount of frequency and imbalance in running server during resource utilization in green cloud. For balancing the load on the Infrastructure as a Service (IaaS), an energy efficient load balancing approach using global load optimization (GC-GLO) method is presented on the green cloud data center in this paper. The energy efficient load balancing approach uses the Global Load Optimization (GLO) method to allocate the resources in a dynamic manner using the load factor in two ways. First, GC-GLO approach uses the logical hierarchy process to allocate the resources in the servers present in green cloud zone. The logical hierarchy process in GC-GLO selects the task in a hierarchical manner and allocates the task between cloud severs’ based on the resource need. With the exact form of resource utilization in GC-GLO, improves the utilization rate in the green cloud environment. Based on the resource utilization, the GC-GLO approach then uses the global load optimization algorithm to measure the load rate of each green cloud server. The energy expenditure on the resource allocation and the IaaS maintenance on the green cloud are eventually measured. The approach, GC-GLO has been validated by conducting careful performance measure study using CloudSim platform in terms of factors such as energy expenditure rate on IaaS maintenance, resource utilization rate and running time. The results demonstrate that global load optimization has extensive potential as it offers significant performance gains to load balancing efficiency and false positive rate under dynamic workload scenarios.