Abstract

This note discusses the application of an almost re- generative method to the simulation of a Markov chain that takes a very long time to converge. Even simulation runs of 10/sup 12/ cells did not produce the same results for simulation scenarios with an average cell-loss-rate of 10/sup -8/. The regenerative method can not be used here as the time between regenerative states is too long (10/sup 30/ events). Instead, an almost regenerative method was tried and it gave results that agreed with predictions from regenerative theory and with some very long simulation runs. The regenerative runs were done on a massively parallel processor.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call