Improvement And Optimization Of Gossip Algorithm Based On Semi-distributed Blockchain Network

Gossip algorithm as an important algorithm in the distributed network system,many blockchain projects based on the semi distributed network use the gossip algorithm to synchronize data between nodes.Although the gossip algorithm is simple,efficient and fault-tolerant,in the actual blockchain network,the original gossip algorithm uses a fixed probability to select the target node to spread the message,so a large number of redundant messages will be generated in the process of data synchronization,and the efficiency of data synchronization will be affected.This paper is based on the related research of traditional gossip algorithm,two improved gossip algorithms are designed for the message propagation model in the semi distributed blockchain network.The main work of this paper is as follows:(1)This paper proposes an improved HNA-Gossip algorithm.A new history node list structure is added to this algorithm.This list is used to store the history nodes that have received messages in the process of dissemination and add them to the sent messages.It is updated with the dissemination.Thus,it can avoid sending duplicate messages to the nodes in the list,reduce the redundancy of messages,and improve the efficiency of data synchronization.(2)In order to solve the problem of poor performance of HNA-Gossip algorithm when there are many propagation branch paths,an improved HMS-Gossip algorithm is proposed.Based on HNA-Gossip algorithm,a state message based broadcast mechanism is added to the algorithm.After receiving the message,the nodes in each branch path actively broadcast the status message to other nodes to indicate the status of the received message,so as to avoid sending duplicate messages to the branch path nodes and further improve the performance of the algorithm.(3)The simulation experiments are carried out on the hyperledger fabric and Anaconda platform respectively.The experimental results show that the performance of the two improved gossip algorithm is better than that of the original gossip algorithm,and HMS-Gossip algorithm is better,which proves the feasibility and effectiveness of the algorithm.