Proactive Problem Avoidance and Quality of Service (QOS) Guarantees for Large Heterogeneous Networks.

摘要

The two primary objectives of this research are: (1) To develop innovative, real-time 'early warning' techniques to detect performance anomalies in large scale, heterogeneous networks that would not be detected by current single element network monitoring/alarm techniques, or to detect those anomalies before current techniques would detect them; and (2) To develop adaptive routing and traffic management methods which enable traffic to proactively avoid these potential problem areas before the problems become serious. This will not only improve the network performance seen by this traffic, but it will also prevent additional traffic from compounding whatever local problems were detected.