OASIS: self-tuning storage for applications

This paper introduces OASIS, a self-tuning storage management architecture that allows applications and the storage environment to negotiate resource allocations with-out requiring human intervention. The goal of OASIS is to maximize the utilization of all the storage resources in the storage area network subject to fairness in the allo-cation of resources to applications. To achieve this goal, OASIS continually inspects the application's I/O behavior, and determines the application's storage requirements au-tomatically. Subsequently, OASIS communicates this in-formation from the applications to a storage manager by means of a communication protocol that isolates the com-plexities of applications and storage from each other. The OASIS framework includes algorithms to dynamically and continuously map application requirements into appropri-ate low-level resource allocations. A key advantage of OA-SIS over best-effort systems is the ability to fairly share storage resources between workloads of varying charac-teristics. What distinguishes OASIS from other managed systems however, is its self-tuning nature. We implemented a prototype of the OASIS architecture and performed ex-periments on a set of competing synthetic workloads de-rived from traces. Our results show that OASIS is able to detect the bandwidth and latency requirements of the com-peting workloads and generate a fairer allocation of stor-age resources than a best-effort approach. More impor-tantly, experience with a real-life database scenario, shows that OASIS is able to satisfy the bandwidth requirements of competing multi-threaded workloads without any storage administrator input. In particular, OASIS is able to iden-tify an under-performing workload and ensure it receives a fair share of the overall storage system resources, resulting in a performance increase by as much as a factor of ve for that workload over best-effort resource allocation. 1