| Andrew Miller, Yu Xia, Kyle Croman, Elaine Shi, Dawn Song
The paper introduces HoneyBadgerBFT, a novel asynchronous Byzantine fault-tolerant (BFT) protocol designed for large-scale, high-transaction-rate applications, particularly in the context of cryptocurrencies and financial transactions. Unlike traditional BFT protocols that rely on weakly synchronous network assumptions, HoneyBadgerBFT guarantees liveness without making any timing assumptions, making it suitable for unpredictable and adversarial network environments.
Key contributions of the paper include:
1. **HoneyBadgerBFT Protocol**: The first practical asynchronous BFT protocol that achieves optimal asymptotic efficiency in atomic broadcast. It uses a novel atomic broadcast protocol based on threshold public-key encryption and efficient reliable broadcast techniques.
2. **Performance**: The protocol demonstrates throughput of tens of thousands of transactions per second and scales to over a hundred nodes on a wide area network, including over Tor without parameter tuning.
3. **Security and Liveness**: Formal proofs of security and liveness are provided, showing that HoneyBadgerBFT outperforms classical PBFT protocols in terms of throughput even under optimistic conditions.
4. **Implementation and Experiments**: A full implementation is provided, and experimental results from an Amazon AWS deployment with over 100 nodes across five continents are presented, along with tests on the Tor network.
The paper also discusses the challenges and limitations of weakly synchronous BFT protocols, particularly in adversarial network environments, and highlights the importance of robustness and decentralization in cryptocurrency applications. HoneyBadgerBFT is positioned as a natural choice for both confederation cryptocurrencies and permissionless blockchains, offering high throughput and robustness in diverse deployment scenarios.The paper introduces HoneyBadgerBFT, a novel asynchronous Byzantine fault-tolerant (BFT) protocol designed for large-scale, high-transaction-rate applications, particularly in the context of cryptocurrencies and financial transactions. Unlike traditional BFT protocols that rely on weakly synchronous network assumptions, HoneyBadgerBFT guarantees liveness without making any timing assumptions, making it suitable for unpredictable and adversarial network environments.
Key contributions of the paper include:
1. **HoneyBadgerBFT Protocol**: The first practical asynchronous BFT protocol that achieves optimal asymptotic efficiency in atomic broadcast. It uses a novel atomic broadcast protocol based on threshold public-key encryption and efficient reliable broadcast techniques.
2. **Performance**: The protocol demonstrates throughput of tens of thousands of transactions per second and scales to over a hundred nodes on a wide area network, including over Tor without parameter tuning.
3. **Security and Liveness**: Formal proofs of security and liveness are provided, showing that HoneyBadgerBFT outperforms classical PBFT protocols in terms of throughput even under optimistic conditions.
4. **Implementation and Experiments**: A full implementation is provided, and experimental results from an Amazon AWS deployment with over 100 nodes across five continents are presented, along with tests on the Tor network.
The paper also discusses the challenges and limitations of weakly synchronous BFT protocols, particularly in adversarial network environments, and highlights the importance of robustness and decentralization in cryptocurrency applications. HoneyBadgerBFT is positioned as a natural choice for both confederation cryptocurrencies and permissionless blockchains, offering high throughput and robustness in diverse deployment scenarios.