Application of Optimal Scheduling in a Quantum Switch: A Practical Case

In this paper, we consider a quantum switch that plays an essential role in quantum networking. It operates as a two-sided queueing system with link-level entanglements generated stochastically and used for request service. As known from recent theoretical findings, the Average Reward Entanglement (ARE) scheduling policy (based on Markov Decision Process) was proposed for a quantum switch, as classical MaxWeight scheduling policy was shown not to be throughput-optimal for a quantum switch. However, detailed practical guidelines of ARE implementation was not proposed. In this paper, we fill this gap by presenting a quantum switch instance with two links and three classes of requests. Moreover, we compare ARE’s behaviour with that of MaxWeight.