Throughput-Optimal Scheduling via Rate Learning

TL;DR

This paper introduces a 'schedule as you learn' approach for network scheduling that learns average rates to make flexible, throughput-optimal decisions, reducing latency compared to traditional max-weight policies.

Contribution

The paper proposes a novel rate learning-based scheduling method that decouples decisions from queue sizes, enabling more flexible and potentially lower-latency network management.

Findings

SYL achieves lower latency for certain flows.

SYL maintains throughput optimality.

Numerical experiments validate the approach.

Abstract

We study the problem of designing scheduling policies for communication networks. This problem is often addressed with max-weight-type approaches since they are throughput-optimal. However, max-weight policies make scheduling decisions based on the network congestion, which can be sometimes unnecessarily restrictive. In this paper, we present a ``schedule as you learn'' (SYL) approach, where we learn an average rate, and then select schedules that generate such a rate in expectation. This approach is interesting because scheduling decisions do not depend on the size of the queue backlogs, and so it provides increased flexibility to select schedules based on other criteria or rules, such as serving high-priority queues. We illustrate the results with numerical experiments for a cross-bar switch and show that, compared to max-weight, SYL can achieve lower latency to certain flows without compromising throughput optimality.