Node-Initiated Byzantine Consensus Without a Common Clock
Danny Dolev, Christoph Lenzen

TL;DR
This paper presents a self-stabilizing Byzantine consensus protocol that operates without a common start time, handling arbitrary clock drifts and message delays within known bounds, improving robustness and efficiency.
Contribution
It introduces a novel framework for node-initiated consensus in semi-synchronous systems that is self-stabilizing and efficient in communication and time.
Findings
Achieves consensus with constant-factor increase in time and communication compared to synchronous start.
Introduces an additive communication overhead of O(n log n) bits per node per time unit.
Framework remains efficient even with arbitrary clock drifts within known bounds.
Abstract
The majority of the literature on consensus assumes that protocols are jointly started at all nodes of the distributed system. We show how to remove this problematic assumption in semi-synchronous systems, where messages delays and relative drifts of local clocks may vary arbitrarily within known bounds. Our framework is self-stabilizing and efficient both in terms of communication and time; more concretely, compared to a synchronous start in a synchronous model of a non-self-stabilizing protocol, we achieve a constant-factor increase in the time and communicated bits to complete an instance, plus an additive communication overhead of O(n log n) broadcasted bits per time unit and node. The latter can be further reduced, at an additive increase in time complexity.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsDistributed systems and fault tolerance · Mobile Agent-Based Network Management · Optimization and Search Problems
