Heavy traffic analysis for EDF queues with reneging (Q535200)

The paper considers a single-server queue in which customers have deadlines and are served until their deadlines elapse. The performance of the system is measured by the fraction of reneged work, defined as the residual work not serviced due to elapsed deadlines. This quantity is shown to be minimized by an earliest-deadline-first scheduling policy. The main result of the paper is a heavy traffic convergence theorem. It is shown that the limit of the scaled workload process is a doubly reflected Brownian motion with lower barrier zero and upper barrier at the mean of the lead time distribution. It is also shown that, conditional on the limiting workload, the resulting limiting measure-valued workload process is the same limiting process as when customers are served to completion.