A Cost-Sensitive Adaptation Engine for Server Consolidation of Multitier Applications

Virtualization-based server consolidation requires runtime resource reconfiguration to ensure adequate application isolation and performance, especially for multitier services that have dynamic, rapidly changing workloads and responsiveness requirements. While virtualization makes reconfiguration easy, indiscriminate use of adaptations such as VM replication, VM migration, and capacity controls has performance implications. This paper demonstrates that ignoring these costs can have significant impacts on the ability to satisfy response-time-based SLAs, and proposes a solution in the form of a cost-sensitive adaptation engine that weighs the potential benefits of runtime reconfiguration decisions against their costs. Extensive experimental results based on live workload traces show that the technique is able to maximize SLA fulfillment under typical time-of-day workload variations as well as flash crowds, and that it exhibits significantly improved transient behavior compared to approaches that do not account for adaptation costs.