Size-based scheduling vs fairness for datacenter flows: a queuing perspective

TL;DR

This paper challenges the effectiveness of SRPT flow scheduling in datacenter networks, advocating for fairness-based approaches and proposing a new virtual fair scheduling algorithm that better handles batch and burst traffic.

Contribution

It demonstrates the non-optimality of SRPT under realistic batch and burst traffic assumptions and introduces a simple, fair scheduling algorithm suitable for high-speed datacenter environments.

Findings

SRPT is non-optimal for batch and burst traffic.

Fairness-based scheduling improves bandwidth sharing.

Proposed virtual fair scheduling is simple and effective.

Abstract

Contrary to the conclusions of a recent body of work where approximate shortest remaining processing time first (SRPT) flow scheduling is advocated for datacenter networks, this paper aims to demonstrate that per-flow fairness remains a preferable objective. We evaluate abstract queuing models by analysis and simulation to illustrate the non-optimality of SRPT under the reasonable assumptions that datacenter flows occur in batches and bursts and not, as usually assumed, individually at the instants of a Poisson process. Results for these models have significant implications for the design of bandwidth sharing strategies for datacenter networks. In particular, we propose a novel "virtual fair scheduling" algorithm that enforces fairness between batches and is arguably simple enough to be implemented in high speed devices.