Duplicated Tasks Elimination for Cloud Data Center Using Modified Grey Wolf Optimization Algorithm for Energy Minimization

Task load balancing is a significant challenge in the cloud environment due to the network's performance in relation to the workload of the cloud machines. It also has a direct impact on how much energy is consumed and, consequently, how much money the cloud provider makes. Customer wants and an application's functionalities are both dynamic. In normal situation tasks allocation make issues like under loaded or overloaded but in some point tasks duplication occur which effect the data center performance. Dynamic tasks allocation in cloud computing improve the tasks allocation and reduce tasks duplication. For that purpose, modified Grey wolf optimization algorithm was design that improve tasks allocation and reduce tasks duplication. Thus, it is necessary to simultaneously achieve several competing goals. The proposed resource allocation technique for cloud computing that is adaptive, multi-objective, teaching–learning based. This method strikes a balance between goals like decreasing makespan, lowering total cost, and improving resource utilization. Additionally, the suggested technique reduces tasks duplication and system imbalance. Comparing the suggested algorithm to different algorithms demonstrated its effectiveness. The simulation results showed that the suggested algorithm may greatly improve resource utilization while reducing the user's total cost and makespan. Future research the proposed approach will examine additional methods that enhance the predication capabilities for cloud data centers.