A Lower Bound on Latency Spikes for Capacity-Seeking Network Traffic

TL;DR

This paper provides a theoretical lower bound on latency spikes caused by sudden capacity drops in capacity-seeking protocols like TCP and QUIC, supported by testbed experiments, highlighting fundamental constraints on low-latency network performance.

Contribution

It derives a fundamental lower bound on latency spikes due to capacity drops and validates it across multiple congestion control algorithms through experiments.

Findings

Latency spike lower bound equals round-trip delay times capacity reduction

The bound holds for DCTCP, BBR, and Cubic algorithms

Implications for designing low-latency network technologies

Abstract

Most Internet traffic is carried by capacity-seeking protocols such as TCP and QUIC. Capacity-seeking protocols probe to find the maximum available throughput from sender to receiver, and, once they converge, attempt to keep sending traffic at this maximum rate. Achieving reliable low latency with capacity-seeking end-to-end methods is not yet entirely solved. We contribute a theoretical analysis to this ongoing discussion. In this work, we derive an expression for the minimum size of the spike in latency caused by a sudden drop in network capacity. Our results highlight a quantifiable and fundamental constraint on capacity-seeking network traffic. When end-to-end capacity is suddenly reduced, capacity-seeking traffic inevitably produces a latency spike. A lower bound on this latency spike can be calculated by multiplying the round-trip delay from the network bottleneck to the source of capacity-seeking traffic by the magnitude of the end-to-end capacity reduction. Testbed experiments show that this bound holds for the DCTCP, BBR, and Cubic congestion control algorithms. Our results have implications for the design of low-latency PHY and MAC-layer technologies because we quantify an important transport-layer consequence of unstable traffic rates.