This paper presents a network model for virtual path (VP) management which considers both the topology and capacity allocation for a set of virtual path connections (VPCs). The objective of the model is to find an optimal VP pool which can be used for on-line network management and control. The solution of the model gives a set of VPCs with assigned capacities which optimise the objective function value for a given network configuration. As an application example, the model is used to find an optimal VP pool for an ATM network which suffers a link failure. A comparison is made between fixed virtual path management (FVPM) and dynamic virtual path management (DVPM), comparing the revenue achieved by each. This illustrates how DVPM adapts the VP pool in a robust fashion to achieve maximum revenue in the face of a link failure.

The authors use adaptive feedback and adaptive feedforward control methodologies to avoid congestion at high server utilisation. This addresses the combined connection admission and flow control problem (generic functions for managing and controlling traffic and congestion in BISDN). Using a novel control concept, based on only two groups of traffic (controllable and uncontrollable) they formulate a problem aimed at high (unity) utilisation of resources while maintaining quality of service at prescribed levels. Bounds on operating conditions are derived, and using simulation they show that high utilisation can be achieved as suggested by the theory, together with robustness for unforeseen traffic connections and disconnections. Even with such a high efficiency and strong properties on the quality of service, the only traffic descriptor required from the user is that of the peak rate of the uncontrollable traffic. >