Elastic Resource Allocation in the Cloud

Abstract

In traditional cloud computing, elastic computing can be provided by increasing or decreasing the number of virtual machines (VM). However, the resource needed by applications change swiftly. This coarse-grained resource allocation approach is not a perfect way to satisfy the demand of the workload in short term. In this paper, we present a fine-grained resource allocation method in virtualized servers. This method uses a real-time control model in order to meet the demands of varying workload by sizing the VMs dynamically. We compare, evaluate, and analyze this approach. Experimental results confirm that this approach guarantees performance and resource efficiency for applications.