An Efficient Resource Allocation Technique for Efficient Virtual Machine Placement Using Local Neighbor Search and Stochastic Random Process

Cloud computing environments efficiently utilize resources by managing diverse cloud users through virtualization techniques. Virtualization enhances resource utilization in Physical Machines (PMs). This paper introduces an efficient virtual machine placement algorithm based on an Improved Genetic Algorithm, which minimizes the Average Resource Idle Time of active PMs, thereby improving overall resource utilization in cloud data centers. Genetic Algorithms are effective optimization methods for achieving high accuracy with fewer iterations. However, they often encounter local optimum problems and quick convergence. The proposed algorithm addresses these issues and is implemented using the CloudSim Plus simulator in a standard cloud environment. Results indicate the algorithm reduces resource waiting time in PMs by approximately 52% compared to other algorithms. Additionally, it minimizes the Average Waiting Time and Average Turnaround Time for Virtual Machine placement. Performance analysis confirms its effectiveness, regardless of the number of Virtual Machine requests or available Physical Machines.