Optimization And Application Of Raft Consensus Mechanism Based On Hyperledger Fabric

As an emerging technology,blockchain has become a very popular technology because of its traceability,tamper-proof,and decentralized characteristics.It has been sought after by all walks of life,especially for data security.It is widely used in fields that require high performance and reliability.As a typical consortium chain and private chain technology,Hyperledger Fabric reforms and optimizes the traditional blockchain model.It has significant advantages in identity management,privacy and confidentiality,efficient processing,smart contracts,module design,etc.,and provides a high degree of Confidentiality,flexibility,flexibility and scalability.Compared with other blockchain systems,the modular design of Hyperledger Fabric supports pluggable consensus modules,and users can choose consensus algorithms according to their needs.As an efficient distributed consensus algorithm,Raft consensus can perfectly fit with Hyperledger Fabric,greatly improving the transaction efficiency of the Hyperledger Fabric blockchain,and is especially suitable for occasions with high data chain efficiency requirements.At the same time,because the Raft consensus only supports crash and fault tolerance,it is slightly insufficient in security,so this article has made some corresponding optimizations for the application scenarios of the Raft consensus.In the era of big data,data as a valuable corporate asset often has security problems such as easy loss,easy tampering,easy forgery,etc.,especially databases that carry core business,often becoming a key target of criminal attacks.Therefore,the auditable and traceable data access has become an important means to ensure database security.Therefore,based on the Hyperledger Fabric blockchain,Raft consensus mechanism and other technologies,this paper studies a safe and efficient method for chaining database access logs,and specifically completes the following work:First of all,because Raft is a typical Leader-based consensus algorithm,in order to improve the fairness and security of the leader election in Raft consensus,by analyzing the variation of the hysteresis difference and the variance of the hysteresis difference of the log application in the Raft consensus mechanism the homogeneity test has constructed a new reputation Value model.The reputation Value model is introduced in the leader election process,and a secure Raft consensus mechanism based on reputation Value is proposed.This mechanism reflects the log writing efficiency of the ordering node and the working status of the node by recording the change in the hysteresis difference of each term of the Raft ordering node.The reputation Value of the node is used to limit the probability of a malicious node being elected as the leader,thereby improving the Raft consensus Overall security.Secondly,in order to ensure the efficiency of data access log on-chain,the paper further proposes an R-P-based database access log on-chain method based on the secure Raft consensus mechanism.This method constructs a parliamentary model based on reputation value.After each leader election is completed,nodes with a reputation value greater than or equal to the average reputation value of previous leaders are included in the list of members of the parliament,and the leader node serves as the chairman of the parliament for this term.On the basis of the parliamentary model,the paper proposes a RP-based transaction load balancing strategy.When the Leader node receives a transaction request from the proxy node,it will allocate part of the request forwarding function to other members of the parliament node,and then the member The nodes forward the request to the crowd nodes that they manage.When there are many Raft nodes,this strategy reduces the load of the Leader node in the process of Raft log replication,improves the efficiency of the Raft consensus mechanism,and further improves the efficiency of database access logs on the chain.Finally,in order to verify the correctness of the database log on-chain method proposed in the paper,an actual Hyperledger Fabric private chain was built,and the performance of the database log on-chain method proposed in the paper,the influence of malicious nodes,the reputation value model,and the incentive mechanism were carried out.Validated analysis.The work of the paper shows that the secure Raft consensus mechanism based on reputation value can improve the security and fairness of the leader election of the ranking module while ensuring the efficiency of the leader election of the Raft ranking module;the RP-based database access log on-chain method can guarantee the ranking Under the premise of the security and fairness of the module Leader election,the efficiency of the database access log on the chain is further improved.