Revenue Recovery for Operational SONET/SDH Networks

Shrinking capital expenditure budgets are forcing telecom service providers to explore network management/optimization tools that extract greater utilization from existing transport infrastructure. It is widely believed that large SONET/SDH networks typically have 20-30% "stranded" resources, both in wasted bandwidth and in untracked hardware. Considering the massive installed base of such networks, industry analysts estimate that achieving only 5% asset recovery would equal $15 billion in revenue savings in North America alone [9]. The traditional solution to recover stranded resources is to collect up-to-date inventory and equipment configuration data (i.e. field data) and use it to reconcile various databases present in a service provider network. This approach is complex and expensive, as it requires a new inventory reconciliation framework and large investment in software, equipment and personnel. In this paper, we present a simpler XML-based approach that uses localized heuristics to uncover stranded resources and configuration errors based on field data alone. Our approach synthesizes higher-level models, like circuits and rings from field data, and analyzes the synthesized data to identify stranded resources and configuration errors. Our approach is not designed to be exhaustive like the traditional approach; instead it provides an operator-friendly solution that focuses on specific, well-known problems in the network to achieve high savings without an overwhelming overhead. Field trial results indicate that our approach is indeed very effective. For example, we found that 10% of the circuits in a large carrier network have stranded bandwidth that can be "recovered", and 33% of 1+1 protection circuits have incorrect configurations impacting service reliability.