Prediction Strategies for Self-Adaptive Behavior in Distributed Systems

The autonomic management of large-scale distributed systems now allows performance improvement, availability, and security, while simultaneously reducing the effort and skills required of system administrators. One way that systems can support these abilities is by relying on a continuous monitoring service to keep track of the states of the targeted systems. However, it is challenging to achieve both scalability and high accuracy when dealing with huge amounts of distributed and time-varying metrics in large-scale production systems. Additionally, most of the monitoring services are designed to provide general resource information and do not consider specific information for higher-level services, lacking important control capabilities. In this context, a dynamic adaptation layer is required, to be able to reactive and proactive deal with detected or predicted conditions. In this paper, we present a prediction architecture developed within the MonALISA monitoring framework, providing methods for estimating future values for different parameters on various periods of time. The predictions are used to enhance the self-adaptive behavior of applications and to minimize continuous monitoring costs. We illustrate our work as an efficient strategy to save power in wireless sensor networks by reducing the data traffic and thus maximizing the network lifetime.