iOptimize: A software capability for analyzing and optimizing connection-oriented data networks in real time

• Published in: Bell Labs technical journal Vol. 9; no. 4; pp. 67 - 81
• Main Authors: Asghar, Naeem, Bhatia, Randeep S., Chandwani, Rakesh, Corcoran, Colin, Hao, Fang, Karwisch, Joseph, Koppol, Pramod V., Lakshman, T. V., Siesta, Michael P., Zlatos, Stephen M.
• Format: Journal Article
• Language: English
• Published: Hoboken IEEE 01.01.2005; Wiley Subscription Services, Inc., A Wiley Company
• ISSN: 1089-7089; 1538-7305
• DOI: 10.1002/bltj.20062

This paper describes a service called iOptimize that analyzes and optimizes service providers' connection-oriented data networks. In these networks, online connection routing is used to set up connections quickly, but the simple path selection scheme and the limited information available for online routing can cause network capacity to be used inefficiently. This may result in the loss or disruption of user data transmissions, leading to a degraded quality of service (QoS) in a network capable of supporting much higher throughput. The objective of iOptimize is to detect and analyze such inefficiencies and to offset them by the occasional rerouting of a selected set of connections, while causing little or no disruption of the existing data services and network operations. By helping service providers support data transport services at the QoS levels desired, iOptimize lets them make the most of their existing network resources and helps them defer capital expenditure. Moreover, by tuning their networks to perform at optimal efficiency, it improves end users' experiences, resulting in lower churn/turnover and reduced maintenance costs.