Abstract
This paper presents a methodology for modeling processing delay of network packets for any protocol stack. Real-time, interactive applications such as desktop conferencing are sensitive to accumulated delay (latency). Our model is generic and can be used to model the processing delay on any processor and for any protocol. This is achieved through the definition of a Language for Processor and network Protocol - LPP, which allows the declaration of both processor and protocol characteristics that are required for our model. On the basis of empirical results we are able to predict certain optimizations for edge routers causing minimal processing latencies, for an overall reduction in network latencies.