Improving the Scalability of Data Center Networks With Traffic-Aware Virtual Machine Placement
The scalability of data center networks has always been a point of concern and has gained a lot of attention in the recent years. A number of researches have been conducted in the past years and continuous efforts have been made to improve existing technology. This paper focuses on the use of traffic-aware Virtual Machine (VM) placement to improve network scalability. As opposed to the traditional methods that require changes in the network architecture and the routing protocols, the new method works by optimizing the placement of VMs on host machines. The traffic patterns among VMs can be better associated with the communication distance by optimization of the VMs, for instance VMs with large mutual bandwidth usage are assigned to host machines in close proximity. The modern data centers are used to formulate VM placement as an optimization problem and prove its hardness. The arrangement runs on a two-tier approximate algorithm that professionally solves the VM placement problem for very large problem sizes. As significant differences have been noticed in the traffic patterns and structures in the past few years, the paper also provides an analysis on the impact of the traffic patterns and network architectures on the potential performance gain of traffic-aware VM placement. Therefore, how VM placement can be effective in improving network scalability under different network topologies and traffic patterns is what the paper focuses on.