BALLAST: Bandit-Assisted Learning for Latency-Aware Stable Timeouts in Raft

TL;DR

BALLAST introduces a bandit-based adaptive timeout mechanism for Raft, improving stability and recovery times under challenging network conditions by replacing static heuristics with contextual learning.

Contribution

The paper proposes BALLAST, a novel online adaptation method using linear contextual bandits to optimize Raft timeouts during network instability.

Findings

Reduces recovery time in WAN scenarios

Decreases unwritable time during network turbulence

Performs competitively on stable networks

Abstract

Randomized election timeouts are a simple and effective liveness heuristic for Raft, but they become brittle under long-tail latency, jitter, and partition recovery, where repeated split votes can inflate unavailability. This paper presents BALLAST, a lightweight online adaptation mechanism that replaces static timeout heuristics with contextual bandits. BALLAST selects from a discrete set of timeout "arms" using efficient linear contextual bandits (LinUCB variants), and augments learning with safe exploration to cap risk during unstable periods. We evaluate BALLAST on a reproducible discrete-event simulation with long-tail delay, loss, correlated bursts, node heterogeneity, and partition/recovery turbulence. Across challenging WAN regimes, BALLAST substantially reduces recovery time and unwritable time compared to standard randomized timeouts and common heuristics, while remaining competitive on stable LAN/WAN settings.