Adaptive Resource Allocation of Multiple Servers for Service-Based Systems in Cloud Computing

Abstract

Due to the advantages of cloud computing, it has been adopted as deployment platform of SBS (Service-based Systems). It also provides an elastic "pay-as-you-go" mode, which creates new resource allocation challenge that satisfying the QoS (Quality of Service) requirements with least resource allocation. There has been much interest in using feedback control to make resource allocation, but these works focus on a single control that does not take the interactions between servers share and compete for the same resource pool. In this paper, we present an adaptive resource allocation approach for SBS in the cloud environment using MIMO (Multi-Input and Multi-Output) control to allocate resource to multiple servers according to multiple workloads. The experimental results show that our approach can ensure the QoS with least resource allocation and increase the resource utilization.