Correlation-Aware Heuristic Search for Intelligent Virtual Machine Provisioning in Cloud Systems

The optimization of resource is crucial for the operation public cloud systems such as Microsoft Azure, well servers dedicated to workloads large customers 365. Those tasks often need take unknown parameters into consideration and can be formulated Prediction+Optimization problems. This paper proposes a new method named Correlation-Aware Heuristic Search (CAHS) that capable accounting uncertainty in delivering effective solutions difficult We apply this solving predictive virtual machine (VM) provisioning (PreVMP) problem, where VM plans are optimized based on predicted demands different types, ensure rapid provisions upon customers' requests pursue high utilization. Unlike current state-of-the-art PreVMP approaches assume independence among CAHS incorporates demand correlation when conducting prediction novel way. Our experiments two benchmarks one industrial benchmark demonstrate achieve better performance than its nine competitors. has been successfully deployed Azure significantly improved performance. main ideas have also leveraged improve efficiency reliability services provided by