🤖 AI Summary
Constructing and maintaining robust overlay topologies in highly dynamic, churn-prone peer-to-peer networks with a large fraction of Byzantine nodes remains an open challenge. Existing approaches rely on global knowledge or strong synchrony assumptions, limiting scalability and practicality.
Method: We propose the first fully distributed, locally bootstrapped protocol for overlay construction and maintenance. It employs randomized neighbor exchange, multi-round sampling-based validation, and lightweight consistency checks—requiring only polylog(n) communication and computation per node per round.
Contribution/Results: With high probability, the protocol maintains a constant-degree expander graph, ensuring network connectivity, O(log n) diameter, and strong scalability. It tolerates up to o(n/polylog n) Byzantine nodes—a novel robustness guarantee—and supports fast Byzantine-tolerant consensus and leader election. Our approach significantly improves upon prior work by eliminating global view requirements and relaxing synchrony assumptions, enabling deployment in realistic, adversarial, high-churn environments.
📝 Abstract
We address a fundamental problem in Peer-to-Peer (P2P) networks, namely, constructing and maintaining dynamic P2P overlay network topologies with essential properties such as connectivity, low diameter, and high expansion, that are resilient to continuous high churn and the presence of a large number of malicious (Byzantine) nodes. Our main goal is to construct and maintain a sparse (bounded degree) expander topology despite high churn and a large number of Byzantine nodes. Such an expander topology has logarithmic diameter, high expansion, and is robust to churn and the presence of a large number of bad nodes, and facilitates efficient and robust algorithms for fundamental problems in distributed computing, such as agreement, broadcasting, routing, etc. Our main contribution is a randomized, fully-distributed dynamic P2P protocol that works with only local initial knowledge and guarantees, with a high probability, the maintenance of a constant degree graph with high expansion even under continuous churn and in the presence of a large number of Byzantine nodes. Our protocol can tolerate up to $o(n/polylog(n))$ Byzantine nodes (where $n$ is the stable network size). Our protocol is efficient, lightweight, and scalable, and it incurs only $O(polylog(n))$ overhead for topology maintenance: only polylogarithmic (in $n$) bits need to be processed and sent by each honest node per round, and any honest node's computation cost per round is also polylogarithmic. Our protocol can be used as a building block for solving fundamental distributed computing problems in highly dynamic networks, such as Byzantine agreement and Byzantine leader election, and enables fast and scalable algorithms for these problems.