An improved multi-objective eagle algorithm for virtual machine placement in cloud environment

In cloud environment, the optimal placement of virtual machines on available physical machines is challenging task. The available resources can be properly managed and the resource wastage can be reduced with the help of optimal placement of virtual machines. The proposed EAGLE modified approach performs better than the existing four optimization approaches namely multi resource alignment, first fit, two stage best fit, EAGLE for optimal placement of virtual machines in cloud computing. These algorithms are simulated using the designed simulator for different sets of workloads. The CPU utilization or power consumption varies from 68 to 88% for first fit, 69–90% for multi resource alignment, 66–87% for two stage best fit, 66–84% for EAGLE, and 66–84% for EAGLE MOD. The memory utilization varies from 40 to 60% for first fit, 50–69% for multi resource alignment, and 38–65% for two stage best fit, EAGLE, and EAGLE MOD. The average resource fragmentation are 30.4388% for first fit, 28.6983% for multi resource alignment, 33.2822% for two stage best fit, and 25.8717% for EAGLE and EAGLE MOD. The average resource fragmentation are 30.4388% for first fit, 28.6983% for multi resource alignment, 33.2822% for two stage best fit, and 25.8717% for EAGLE and EAGLE modified.The presented algorithms are compared on the basis of central processing unit (CPU) utilization, memory utilization, failure tasks, and resource fragmentation evaluation parameters.