RPPS: A Novel Resource Prediction and Provisioning Scheme in Cloud Data Center

Abstract

Cloud data centers and virtualization are being highly considered for enterprises and industries. However, elastic fine-grained resource provision while ensuring performance and SLA guarantees for applications requires careful consideration of important and extremely challenging tradeoffs. In this paper, we present RPPS (Cloud Resource Prediction and Provisioning scheme), a scheme that automatically predict future demand and perform proactive resource provisioning for cloud applications. RPPS employs the ARIMA model to predict the workloads in the future, combines both coarse-grained and fine-grained resource scaling under different situations, and adopts a VM-complementary migration strategy. RPPS can resolve predictive resource provisioning problem when enterprises confront demand fluctuations in cloud data center. We evaluate a prototype of RPPS with traces collected by ourselves using typical CPU intensive applications and as well as workloads from a real data center. The results show that it not only has high prediction accuracy (about 90% match in most time) but also scales the resource well.