It is significant to improve Practical Byzantine Fault Tolerance algorithm (PBFT) in consortium blockchain. At present, the serial verification process of transactions in the primary and backups greatly affects consensus efficiency. Meanwhile, the lack of reasonable valuation mechanism in PBFT makes it difficult to motivate existing reliable nodes. Moreover, consensus nodes work in an enclosed environment, where nodes cannot join and exit dynamically. To solve the shortcomings stated above, we propose a dynamic hierarchical Byzantine fault-tolerant consensus mechanism based on credit (DHBFT). Firstly, we design a hierarchical-parallel scheme composed of consensus nodes, candidate nodes, and ordinary nodes. We realize parallel transaction logic verification in the primary and backups by delegating candidate nodes to verify the validity of transactions preliminarily. Secondly, we create a reward-punishment scheme. The consensus nodes with better performances are assigned higher credit value and have higher probability to become the primary. Thirdly, we propose a dynamic promotion-demotion scheme. It enables faulty nodes to be excluded from the consensus set and reliable candidate nodes to join. Experimental results show that DHBFT has better efficiency and higher stability. Compared with PBFT, the overall throughput of transactions is increased by 16%, and the average delay is reduced by 12%. Moreover, the proportion of abnormal nodes is basically 0 and much lower than that of PBFT.
[1]
Miguel Castro,et al.
Practical byzantine fault tolerance and proactive recovery
,
2002,
TOCS.
[2]
Satoshi Nakamoto.
Bitcoin : A Peer-to-Peer Electronic Cash System
,
2009
.
[3]
Li Zhang,et al.
Research on Consensus Efficiency Based on Practical Byzantine Fault Tolerance
,
2018,
2018 10th International Conference on Modelling, Identification and Control (ICMIC).
[4]
Yuhao Wang,et al.
Study of Blockchains’s Consensus Mechanism Based on Credit
,
2019,
IEEE Access.
[5]
Iddo Bentov,et al.
Proof of Activity: Extending Bitcoin's Proof of Work via Proof of Stake [Extended Abstract]y
,
2014,
PERV.
[6]
Rüdiger Kapitza,et al.
Hypervisor-Based Efficient Proactive Recovery
,
2007,
2007 26th IEEE International Symposium on Reliable Distributed Systems (SRDS 2007).