Policies for Assisted Virtual Machine Selection in Cloud Computing Environments

In this paper we propose an end-to-end approach to the VM allocation problem using policies based uniquely on round-trip time measurements between VMs. We propose and implement end-to-end algorithms for VM selection that cover desirable profiles of communications between VMs in distributed applications in a cloud setting. The use of informed VM selection allowed us to select instances with an average RTT up to 85% lower, in one case, and to find local and global centers of VM clusters in another. The approach described is completely independent from cloud architecture and is adaptable to different types of applications and workloads, as well as lightweight and transparent for cloud providers.