A technique based on importance sampling is presented for extrapolating the end-to-end delay performance of a high-level data link control (HDLC) network. The lost packets can arise from bit errors or discarding due to network congestion. The technique works well for a substantial range of loads, as long as the desired performance is at an error rate less than the observed rate (the environment most frequently encountered). It is applicable across a range of error rates. The technique was tested on a simple HDLC link but would generalize to other link-level protocols. The extrapolation technique can be based on simulation or measurement. Simulation results are presented to demonstrate its effectiveness.>

The statistical nature of Monte Carlo techniques presents significant computational problems in the simulation of computer communications networks. A variety of techniques has been developed over the years to improve the computational efficiency of simulation models of general queuing networks, some of which have been applied to communication networks. The authors review the status of efficiency-enhancing techniques as they relate to the simulation of computer communications networks, present some novel approaches, and suggest promising areas for future research.>