Cosine elastic cluster scheduling algorithm for VM migration in cloud computing.

Cloud computing has always been an area of interest for the researchers and the industry oriented professionals. The cloud computing has its primary concern to deliver better performance with low energy consumption and least SLA violations. To rescue the rising challenges in the cloud computing, virtualization is an important process when the Physical Machines (PMs) are expected to work at a higher speed. This paper represents two co-operative algorithms: Cosine Elastic Clustered Scheduling Algorithm (CESCA) and Threshold-Modified Best Fit Algorithm (T-MBFD) for the VM selection and reallocation. CESCA clusters the VMs to be migrated based on three attributes namely CPU-Utilization (CPU), Associated RAM (AsRAM) and Bandwidth Utilization (BU) and the co-relation between them using cosine similarity. CESCA also prioritize the created clusters for the allocation. T-MBFD uses a logarithmic scale to neutralize the priority and the density of the created cluster to finally set the VM to its best suitable PM. The evaluation of the proposed algorithm is done using power consumption and total number of migrations. [ABSTRACT FROM AUTHOR]