Internet router outage measurement: an embedded approach
Sejun Song, Jeff Huang · 2004
Outage measurement is an integral part of high-availability network operations to assess and report the availability of router components and, in turn, the availability of the network. The paper presents a novel approach to outage measurement, called component outage on-line (COOL) measurement. COOL provides an autonomous real-time outage measurement within the router. It automates the outage measurement process and makes it more accurate, reliable, scalable, and cost-effective. The paper describes COOL's measurement methodology with respect to outage model, measurement metrics, architectural framework, methods for measuring hardware and software outages and planned and unplanned outages, and outage MIB (management information base) design. COOL is being implemented in various network routers. Experiment results on COOL runtime performance are presented.