Perceptive VM Allocation in Cloud Data Centers for Effective Resource Management

Virtual Machine allocation in cloud computing centers has become an important research area. Efficient VM allocation can reduce power consumption and average response time which can benefit both the end users as well as the cloud vendors. This work presents a perceptive priority aware VM allocation policy named P-PAVA algorithm, which takes into account the priority of an application along with its compute, memory and bandwidth requirement. The algorithm performs allocation of the applications based on the priority it gets using a machine learning based prediction model. Furthermore, to reduce the overhead of the allocation algorithm, parallelization is employed before assigning various workloads. To achieve this, the algorithm employs the First fit technique as a baseline for the requests allocation with a criteria as low priority. When compared to the state of the art algorithm for VM allocation for priority aware applications, P-PAVA performs better on several criteria such as average response time, execution time and power consumption.