Improved PBFT algorithm for high-frequency trading scenarios of alliance blockchain

Area of Science:

• Blockchain Technology
• Distributed Systems
• Network Security

Background:

• Alliance blockchains require efficient consensus algorithms for scalability.
• The standard Practical Byzantine Fault Tolerance (PBFT) algorithm faces performance limitations with over 100 nodes.
• Existing PBFT has high communication overhead and lacks dynamic node management.

Purpose of the Study:

• To propose an improved consensus algorithm, tPBFT, for high-frequency trading in alliance blockchains.
• To address the scalability and performance issues of PBFT in large-scale networks.
• To enhance the efficiency and dynamic management of consensus nodes.

Main Methods:

• Introduced a trust equity scoring mechanism for dynamic consensus node adjustment.
• Simplified the pre-prepare stage of the PBFT consensus process.
• Implemented transaction list hash verification in the reply stage to reduce overhead.

Main Results:

• The tPBFT algorithm demonstrates superior performance compared to PBFT for networks with more than 30 nodes.
• Significant improvements observed in node communication overhead, consensus efficiency, and scalability.
• tPBFT effectively manages dynamic node adjustments through its trust-based scoring system.

Conclusions:

• The tPBFT algorithm offers a viable solution for enhancing alliance blockchain performance in high-frequency trading.
• tPBFT overcomes the limitations of traditional PBFT, providing better scalability and efficiency.
• The trust-based approach in tPBFT enables dynamic consensus node management, crucial for evolving networks.