Problem
Research questions and friction points this paper is trying to address.
Lack of formal framework for RSM communication
Need efficient protocol for cross-cluster communication
Improving performance in fault-tolerant consensus protocols
Picsou: Enabling Efficient Cross-Consensus Communication
🤖 AI Summary
Existing cross-consensus cluster (RSM) communication lacks an efficient, general-purpose protocol framework, resulting in high interaction overhead and poor reliability among heterogeneous fault-tolerant systems. This paper proposes C3B, a broadcast primitive, and its implementation PICSOU. C3B introduces the first precise message-state determination mechanism based on quorum acknowledgements (QUACKs) and pioneers the adaptation of TCP-inspired heuristic retransmission strategies to cross-RSM communication—achieving constant metadata overhead while preserving strong fault adaptivity. By abstracting consensus interfaces, PICSOU uniformly supports crash-stop and Byzantine fault-tolerant (BFT) models, enabling interoperability across heterogeneous RSM protocols. Experimental evaluation shows that, compared to all-to-all broadcasting, C3B achieves up to 24× higher throughput; significantly reduces retransmissions under failures; and maintains strictly constant metadata overhead in failure-free operation.
📝 Abstract
Replicated state machines (RSMs) cannot effectively communicate today as there is no formal framework or efficient protocol to do so. To address this issue, we introduce a new primitive, the Cross-Cluster Consistent Broadcast (C3B) and present PICSOU, a practical C3B implementation. PICSOU draws inspiration from networking and TCP to allow two RSMs to communicate with constant metadata overhead in the failure-free case and minimal number of message resends in the case of failures. PICSOU is flexible and allows both crash fault-tolerant and byzantine fault-tolerant protocols to communicate. At the heart of PICSOU's good performance and generality lies a novel technique we call QUACKs (quorum acknowledgements) that allow nodes in each RSM to precisely determine when messages have definitely been received, or definitely been lost. Our results are promising: we obtain up to 24x better performance than existing all-to-all solutions.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Introduces Cross-Cluster Consistent Broadcast (C3B)
Uses QUACKs for precise message tracking
Achieves 24x better performance than predecessors
🔎 Similar Papers
No similar papers found.
