Adaptive Address Family Selection for Latency-Sensitive Applications on Dual-stack Hosts

TL;DR

This paper investigates address family selection on dual-stack hosts to reduce latency for internet applications, proposing a dynamic, learning-based mechanism validated through measurements and real-world tests.

Contribution

It introduces a novel online learning approach for dynamic address family selection that adapts per destination to optimize latency.

Findings

The proposed method converges to the lowest-latency address family.

It improves application transport connection latency.

Validated through simulations and real-world experiments.

Abstract

Latency is becoming a key factor of performance for Internet applications and has triggered a number of changes in its protocols. Our work revisits the impact on latency of address family selection in dual-stack hosts. Through RIPE Atlas measurements, we analyse the address families latency difference and establish two requirements based on our findings for a latency-focused selection mechanism. First, the address family should be chosen per destination. Second, the choice should be able to evolve over time dynamically. We propose and implement a solution formulated as an online learning problem balancing exploration and exploitation. We validate our solution in simulations based on RIPE Atlas measurements, implement and evaluate our prototype in four access networks using Chrome and popular web services. We demonstrate the ability of our solution to converge towards the lowest-latency address family and improve the latency of transport connections used by applications.