![]() | Only 14 pages are availabe for public view |
Abstract Software aging has recently gained high momentum due to the increasing demand on high availability and high performance for software systems. Unplanned computer system outages are more likely to be the result of software failures than of hardware failures. Moreover, software often exhibits an increasing failure rate over time, typically because of increasing and unbounded resource consumption, data corruption, and numerical error accumulation. Software rejuvenation is a proactive technique which is used to counteract software aging by terminating the software applications, cleaning their internal state and/or their environment, and then restarting them again. We are concerned with when to schedule software rejuvenation and get the optimal software rejuvenation schedules which maximize the system availability. We propose a proactive technique by using Artificial Neural Networks (ANN) to mine and predict patterns in software aging phenomenon. We analyze resource usage data collected on a typical long-running software system; a web server. A Multi-Layer Perceptron feed forward Artificial Neural Network was trained on an Apache web server dataset to predict future server resource exhaustion through univariate time series forecasting and multivariate ANN empirical modeling techniques. |