Dumbo-NG: Fast Asynchronous BFT Consensus with Throughput-Oblivious Latency

TL;DR

Dumbo-NG targets practical asynchronous BFT by removing the throughput-latency trade-off in atomic broadcast. It achieves this via a direct reduction of asynchronous ABC to multi-valued validated Byzantine agreement (MVBA) with a quality property, enabling completely concurrent transaction dissemination and asynchronous agreement. The approach yields near-constant latency as throughput scales and guarantees censorship resilience without heavy cryptography or memory blow-up. Implemented and evaluated on WAN-scale deployments, Dumbo-NG demonstrates substantially higher peak throughput and stable latency compared with state-of-the-art baselines, signaling a practical path for censorship-resilient, high-throughput asynchronous consensus in geographically distributed settings.

Abstract

Despite recent progresses of practical asynchronous Byzantine fault tolerant (BFT) consensus, the state-of-the-art designs still suffer from suboptimal performance. Particularly, to obtain maximum throughput, most existing protocols with guaranteed linear amortized communication complexity require each participating node to broadcast a huge batch of transactions, which dramatically sacrifices latency. Worse still, the f slowest nodes' broadcasts might never be agreed to output and thus can be censored (where f is the number of faults). Implementable mitigation to the threat either uses computationally costly threshold encryption or incurs communication blow-up, thus causing further efficiency issues. We present Dumbo-NG, a novel asynchronous BFT consensus (atomic broadcast) to solve the remaining practical issues. Its technical core is a non-trivial direct reduction from asynchronous atomic broadcast to multi-valued validated Byzantine agreement (MVBA) with quality property. Most interestingly, the new protocol structure empowers completely concurrent execution of transaction dissemination and asynchronous agreement. This brings about two benefits: (i) the throughput-latency tension is resolved to approach peak throughput with minimal increase in latency; (ii) the transactions broadcasted by any honest node can be agreed to output, thus conquering the censorship threat with no extra cost. We implement Dumbo-NG and compare it to the state-of-the-art asynchronous BFT with guaranteed censorship resilience including Dumbo (CCS'20) and Speeding-Dumbo (NDSS'22). We also apply the techniques from Speeding-Dumbo to DispersedLedger (NSDI'22) and obtain an improved variant of DispersedLedger called sDumbo-DL for comprehensive comparison. Extensive experiments reveal: Dumbo-NG realizes better peak throughput performance and its latency can almost remain stable when throughput grows.

Figures (15)

• Figure 1: Execution flow of an epoch in $\mathsf{HBBFT}$, $\mathsf{Dumbo}$ and their variants. The protocols proceed by consecutive epochs.
• Figure 2: Latency breakdown of $\mathsf{Dumbo}$ (on 16 Amazon EC2 c5.large instances across different regions). $|B|$ is batch size, i.e., the number of tx to broadcast by each node (where each tx is 250-byte to approximate the size of Bitcoin's basic tx). TPKE is a technique from $\mathsf{HBBFT}$ for preventing censorship.
• Figure 4: Execution flow of $\mathsf{DispersedLedger}$. Transaction retrieval is executed concurrently to dispersal and agreement. Modules in the light-orange region represents one $\mathsf{ACS}$.
• Figure 5: High-level of $\mathsf{Dumbo\textrm{-}NG}$. Each node leads an ever-running multi-shot broadcast to disseminate its input transactions. Aside from broadcasts, a sequence of asynchronous multi-valued validated Byzantine agreements ($\mathsf{MVBA}$s) are executed to totally order all broadcasted transactions.
• Figure 6: Latency of dispersal/retrieval as batch size grows.

Theorems & Definitions (7)

• definition 1
• definition 2
• definition 3
• lemma 1
• lemma 2
• lemma 3
• theorem 1