Averaging over fast variables in the fluid limit for Markov chains: Application to the supermarket model with memory (Q1950259)

The paper considers Markov chains having some components which oscillate rapidly and randomly, while others are close to deterministic. It describes a general framework to allow the incorporation of averaging over fast variables into fluid limit estimates for Markov chains. The main results of Section 1 establish explicit error probabilities for the fluid approximation under assumptions which can be verified from knowledge of the transition rates of the Markov chain. Section 2 contains an application of the general theory to the supermarket model with memory.