A survey on scalable consensus algorithms for blockchain technology

The process of reaching an agreement on a value within a distributed network, known as a consensus problem, is a defining feature of blockchain. This consensus problem can be seen in various applications like load balancing, transaction validation in blockchain, and distributed computing. In recent years, many researchers have provided solutions to this problem. Hence we have presented a survey in which we delved into blockchain consensus algorithms and conducted a comparative analysis of all the consensus algorithms to provide information about each protocol’s advantages and drawbacks. This survey starts with the standard proof-of-work consensus protocol applied in bitcoin cryptocurrency and its limitations on the ground of the following parameters: throughput (transactions per second), latency, forks, fault tolerance, double spending attacks, and power consumption. The rest of the consensus algorithms in this paper have been systematically covered to address the limitations of proof-of-work. This paper also covered Raft and PBFT consensus algorithms suitable for permissioned networks. Although the PBFT consensus protocol has a high throughput and a low latency, it has limited node scalability. The PBFT has a low byzantine fault tolerant rate. This paper also covers PoEWAL for blockchain-based IoT applications and WBFT, which prevents corrupt nodes from taking part in consensus. A comparative analysis of the consensus algorithms provides an explicit knowledge of the present research, which also offers guidance for future study.