A resource allocation strategy in cloud using Roulette Wheel Selection method

In recent years cloud computing emerged as the next generation technology by offering numerous services over the internet in resource management. Resource allocation in cloud environment is a challenging task, since there is a disparity between the number of requests for the services and the number of available resources. Hence an effective mechanism is essential to evenly share the load among the available resources. A technique based on Roulette Wheel Selection method is proposed to map the virtual machines (VMs) to physical machines (PMs). The Roulette Wheel is divided into segments based on the probability of assigning VMs to a PM. Fitness values of the individual resources are computed and resource allocation is done accordingly. The performance of the proposed algorithm is evaluated based on the utilization rate of the physical machines. The proposed method shows notable improvements when compared to the First Come First Serve (FCFS) and Stochastic Hill Climbing (SHC) algorithms.