Abstract

One of the challenging problems in cloud computing is the efficient placement of virtual machines having varying resource requirements over the physical machines of fixed capacity. The placement of virtual machines can be done using different approaches. One of the approaches is reducing virtual machine placement problem to multidimensional bin packing problem which is a NP-Hard problem. As the VM placement is a multidimensional problem, it is necessary to consider multiple dimensions such as CPU usage, memory, network bandwidth, etc. together while allocating virtual machines to physical machines. By considering the memory and network affinity among virtual machines during the VM placement, we can significantly reduce the memory usage and network overhead. Existing algorithms do not consider these affinities together while performing the VMs’ placement. In this paper we proposed a multidimensional affinity aware VM placement algorithm which can optimize the performance as compare to the existing algorithms.

Keyword

Cloud computing, virtualization, page sharing, memory affinity, network affinity, VM placement.

Cite this article

Refference