Research On Blockchain Consensus Algorithm For Data Sharin

In today’s era of big data,the society is developing rapidly,and the amount of information circulation is increasing rapidly.Data sharing system can carry out more efficient data transmission and facilitate the function of data realization.Due to the different research objectives and information completeness of the participating institutions,it is impossible to establish efficient interconnection and resource sharing among the systems,and the phenomenon of data island is still quite severe.Blockchain technology is a hot research at present,which has the characteristics of traceability,tamper-proof and decentralization,and has a high compatibility with data sharing system.Blockchain technology ensures the security of data sharing systems,and the consensus algorithm in blockchain is the key to improving performance.However,existing consensus algorithms such as proof-of-work are faced with low throughput,high latency and high computing resources.Therefore,in order to solve the bottleneck problem in the data sharing system,this paper makes an in-depth study of the blockchain consensus algorithm and proposes more efficient consensus algorithm under the premise of guaranteeing the security of blockchain network,namely Improved Po A consensus algorithm and Honey Badger BFT consensus algorithm.It aims to improve the performance of distributed data sharing system.The main research work of this paper is as follows.1.Based on the Proof of Authority(Po A)consensus algorithm(hereafter referred to as Po A consensus algorithm),this paper proposes the Improved Po A consensus mechanism.The “block waiting queue” in the Improved Po A consensus algorithm is mainly used for parallel execution of block generation and verification process to improve the efficiency of consensus.In the Improved Po A consensus algorithm,the verification nodes are reasonably divided into multiple verification groups,and each group of verification nodes verifies different blocks of data transactions in parallel,so that a large number of blocks to be verified are verified at the same time,thus improving the efficiency of consensus and reducing latency.2.This paper presents a more efficient asynchronous Byzantine consensus algorithm,Hipro-Honey Badger BFT,based on the Honey Badger BFT consensus algorithm,which uses a parallel operation mechanism and decouples the ACS protocols by adding several staging pools to the ERBC and ABA protocols and selecting some ERBC instances in the reliable broadcast protocols.This reduces the number of subsequent ABA instances,reduces the amount of communication between nodes during the agreement process of the ACS protocol,and optimises the computational resources in the algorithm.The algorithm is suitable for asynchronous environments and divides the consensus algorithm process into two regions,so that the broadcast process and the consensus process can be executed in parallel,avoiding their interlocking.The algorithm reduces the computational resources of the system and improves the consensus efficiency of the algorithm without sacrificing security.3.The Improved Po A and Hipro-Honey Badger BFT consensus algorithms proposed in this paper are verified and evaluated.The experimental results in this paper show that the proposed Improved Po A consensus algorithm improves the throughput by 27.65% and reduces the latency by 21.66% over the Po A consensus algorithm without sacrificing system security.Utilisation was reduced by 54.09%.