Matrix: Achieving Predictable Virtual Machine Performance in the Clouds

The success of cloud computing builds largely upon on-demand supply of virtual machines (VMs) that pro-vide the abstraction of a physical machine on shared re-sources. Unfortunately, despite recent advances in virtu-alization technology, there still exists an unpredictable performance gap between the real and desired perfor-mance. The main contributing factors include contention to the shared physical resources among co-located VMs, limited control of VM allocation, as well as lack of knowledge on the performance of a specific VM out of tens of VM types offered by public cloud providers. In this work, we propose Matrix, a novel performance and resource management system that ensures the de-sired performance of an application achieved on a VM. To this end, Matrix utilizes machine learning methods-clustering models with probability estimates- to predict the performance of new workloads in a virtualized en-vironment, choose a suitable VM type, and dynamically adjust the resource configuration of a virtual machine on the fly. The evaluations on a private cloud, and two pub-lic clouds (Rackspace and Amazon EC2) show that for an extensive set of cloud applications, Matrix is able to estimate application performance with average 90 % ac-curacy. In addition, Matrix can deliver the target per-formance within 3 % variance, and do so with the best cost-efficiency in most cases.