Autonomic energy management of multi-tier clustered applications

Abstract

Nowadays, medium or large-scale distributed infrastructures such as clusters and grids are widely used to host various kinds of applications (e.g. web servers or scientific applications). Power consumption is becoming a major challenge for most organizations that run these infrastructures. Many studies show that they are not used at their full capacity and that there are therefore a huge source of wasted power.Autonomic management systems have been introduced in order to dynamically adapt software infrastructures according to runtime conditions. They provide support to deploy, configure, monitor, and repair applications in such environments.In this paper, we report our experiments in using Tune -an autonomic management system -to provide energy aware management for a clustered J2EE application. We use Tune to dynamically adapt the degree of replication of the J2EE tiers in the cluster, and to dynamically turn cluster nodes on -to handle the load when it raises up -and off -to save power under lighter load.

Energy: A New Criteria for Performances in Large Scale Distributed Systems

“…In these works, the applications state their performance to achieve or the energy cap not to exceed. For instance, [3] uses machine learning techniques to achieve SLA specifications while [17] use autonomic computing for the same.…”