Classification and Performance Comparative Analysis of Distributed Consensus Algorithms

Yilan Rao

doi:10.54097/m9dpge78

Authors

Yilan Rao School of Dundee International College, Central South University, Changsha, 410083, China

DOI:

https://doi.org/10.54097/m9dpge78

Keywords:

Paxos, Raft, PBFT, Blockchain, Comparison of consensus algorithm performance.

Abstract

Distributed consensus algorithms are the core mechanism for ensuring data consistency and fault tolerance in distributed systems. Their evolution, from classical algorithms like Paxos and Raft to blockchain, reflects the diverse requirements for consistency, security, and performance in different application scenarios. This article analyzes the classification system of distributed consensus algorithms and divides them into two major categories based on fault models: Byzantine fault-tolerant and non-Byzantine fault-tolerant. On this basis, the system systematically organized the processes and characteristics of Paxos, Raft, Practical Byzantine Fault Tolerance (PBFT), and blockchain consensus algorithms, and performed a comparative analysis of two key performance indicators: throughput and consensus latency. Research has found that as the number of nodes increases, the performance of the traditional PBFT algorithm decreases significantly, while the Byzantine Fault Tolerant (TBFT) algorithm shows better scalability in both throughput and latency. Finally, this article combines the characteristics of various algorithms, summarizes their applicable scenarios, and provides theoretical basis and practical reference for the selection of consensus algorithms in distributed systems.

References

[1] Challa S J, Goyal N, Sharma A, et al. A Survey and Experimental Review on Data Distribution Strategies for Parallel Spatial Clustering Algorithms. Journal of Computer Science and Technology, 2024, 39(3): 610-636.

[2] Shuo W, Yongcai W, Deying L, et al. Distributed Relative Localization Algorithms for Multi-Robot Networks: A Survey. Sensors, 2023, 23(5): 2399-2399.

[3] Dezhi M, Wei W, Leiming S, et al. Application and Performance Advantage Verification of Distributed Computing in Petroleum Exploration Data Processing, 2025, 52(12): 22-24.

[4] Chunzhang L, Lele M, Da O, et al. Research on Optimization and Application of the Paxos Algorithm. Internet of Things technology, 2025, 15(17): 116-118.

[5] Jingxiong G. Optimization Migration Method Based on TLA Specification Optimization Migration Method Based on TLA Specification. Jiangxi: Nanchang University, 2023.

[6] Miao T. Research and Application of the Raft Consensus Algorithm. Henan: Zhengzhou University, 2022.

[7] Fanyao M, Xinxin L, Nan F, et al. Pipeline Log Replication Method Optimized Based on the Raft Consensus Algorithm. Telecommunication Engineering Technology and Standardization, 2025, 38(02): 41-46.

[8] Sheping Z, Chaoyue K, Rui Y, et al. RC-PBFT: An Improved PBFT Algorithm Based on Reputation Grouping. Computer Engineering, 2026, 1-12.

[9] Yong L. Research on Blockchain Consensus Algorithms Based on Evolutionary Game Theory. Jiangxi: Jiangxi University of Science and Technology, 2025.

[10] Yong X, Chuanheng S, Na L, et al. Improved Byzantine Fault Tolerant Consensus Algorithm for the Internet of Things. Computer Engineering and Design, 2025, 46(02): 360-367.