Efficient failure prediction in autonomic networks based on trend and frequency analysis of anomalous patterns

We describe an efficient failure prediction system based on new algorithms that model and detect anomalous behaviors using multi-scale trend analysis of multiple network parameters. Our approach enjoys many advantages over prior approaches. By operating at multiple timescales simultaneously, the new system achieves robustness against unreliable, redundant, incomplete and contradictory information. The algorithms employed operate with low time complexity, making the system scalable and feasible in real-time environments. Anomalous behaviors identified by the system can be stored efficiently with low space complexity, making it possible to operate with minimal resource requirements even when processing high-rate streams of network parameter values. The developed algorithms generate accurate failure predictions quickly, and the system can be deployed in sa distributed setting. Prediction quality can be improved by considering larger sets of network parameters, allowing the approach to scale as network complexity increases. The system is validated by experiments that demonstrate its ability to produce accurate failure predictions in an efficient and scalable manner.