Dynamic Server Allocation Under Stochastic Switchover on Time-Varying Links

TL;DR

This paper introduces ACI, a novel non-myopic scheduling framework that effectively manages stochastic switching delays in dynamic server allocation, improving throughput and latency trade-offs in time-varying network links.

Contribution

The paper proposes ACI, a new frame-based scheduling policy that accounts for stochastic switching delays, proving its throughput optimality and demonstrating its effectiveness in UAV network scenarios.

Findings

ACI is throughput-optimal under scaled capacity.

ACI improves network throughput in multi-UAV FSO backhaul.

Adapting urgency metrics balances throughput and latency.

Abstract

Dynamic resource allocation to parallel queues is a cornerstone of network scheduling, yet classical solutions often fail when accounting for the overhead of switching delays to queues with superior link conditions. In particular, system performance is further degraded when switching delays are stochastic and inhomogeneous. In this domain, the myopic, Max-Weight policy struggles, as it is agnostic to switching delays. This paper introduces ACI, a non-myopic, frame-based scheduling framework that directly amortizes these switching delays. We first use a Lyapunov drift analysis to prove that backlog-driven ACI is throughput-optimal with respect to a scaled capacity region; then validate ACI's effectiveness on multi-UAV networks with an FSO backhaul. Finally, we demonstrate how adapting its core urgency metric provides the flexibility to navigate the throughput-latency trade-off.