iScale: An Intelligent Auto-Scaling Engine for Migrated Applications to the Cloud
Omid Bushehrian, Zahra Reisi · 2014
Migration of the legacy software systems to cloud environment is a very cost-effective solution for the service providers. It is primarily due to the elastic nature of the cloud environment: The migrated software is able to acquire and release resources on-demand based on the incoming workload and by doing this not only the service provider has to pay only for the resources they need, but also the QoS requirements of the end-users are always met. To achieve this, the migrated software has to be supported with an “Auto-Scaling Engine ” to add or remove resources to/from it at runtime based on the incoming workload. However, Reacting to the transient workload changes always results in useless sequences of “acquire-release ” actions of cloud resources which impose unwanted overhead costs on the service provider due to the configuration changes. In this paper an “Intelligent Auto-Scaling Engine ” which recognizes the transient workload changes using a Learning Automata and only reacts to the stable ones is proposed. We observed that the proposed method is very effective in reducing the overall costs of the service provider.