Using Ant Colony System to Consolidate Multiple Web Applications in a Cloud Environment

Infrastructure as a Service (IaaS) clouds provide virtual machines (VMs) under a pay-per-use business model, which can be used to create a dynamically scalable cluster of servers to deploy one or more web applications. In contrast to the traditional dedicated hosting of web applications where each VM is used exclusively for one particular web application, the shared hosting of web applications allows improved VM utilization by sharing VM resources among multiple concurrent web applications. However, in a shared hosting environment, dynamic scaling alone does not minimize over-provisioning of VMs. In this paper, we present a novel approach to consolidate multiple web applications in a cloud-based shared hosting environment. The proposed approach uses Ant Colony Optimization (ACO) to build a web application migration plan, which is then used to minimize over-provisioning of VMs by consolidating web applications on under-utilized VMs. The proposed approach is demonstrated in discrete-event simulations and is evaluated in a series of experiments involving synthetic as well as realistic load patterns.