Abstract
Self-healing telecommunications networks within AT&T are the bandwidth-management service (BMS) and bandwidth management service-extended (BMS-E). The authors summarize a self-healing network's capabilities and architecture, using BMS-E as an example. Error performance and availability objectives related to the self-healing characteristics are described. Appropriate physical and network control program (NCP) parameters needed to maintain IBM systems network architecture (SNA) sessions in self-healing network environments are given. Guidelines and examples of NCP parameter values set to maximize the availability and reliability of SNA sessions are presented. The impact of a self-healing network on IBM SNA application performance is discussed, and it is demonstrated that introduction of self-healing capabilities into an SNA environment can substantially increase application availability and reliability as long as network design guidelines are respected.<>