Responding rapidly to service level violations using virtual appliances

One of the key goals in the data center today is provid-ing storage services with service-level objectives (SLOs) for performance metrics such as latency and throughput. Meet-ing such SLOs is challenging due to the dynamism observed in these environments. In this position paper, we propose dynamic instantiation of virtual appliances, that is, virtual machines with storage functionality, as a mechanism to meet storage SLOs efficiently. In order for dynamic instantiation to be realistic for rapidly-changing environments, it should be automated. Therefore, an important goal of this paper is to show that such au-tomation is feasible. We do so through a caching case study. Specifically, we build the automation framework for dynam-ically instantiating virtual caching appliances. This frame-work identifies sets of interfering workloads that can bene-fit from caching, determines the cache-size requirements of workloads, non-disruptively migrates the application to use the cache, and warms the cache to quickly return to accept-able service levels. We show through an experiment that this approach addresses SLO violations while using resources ef-ficiently. 1