An Architecture Evolution Algorithm For Fabric Blockchain Application Software

Hyperledger Fabric is a mainstream enterprise-level blockchain platform.It is popular with enterprise customers for its openness and wide applicability.The nodes of the Fabric system are mainly divided into Peer nodes and Orderer nodes.Peer nodes are responsible for sending transactions,Orderer nodes are responsible for receiving transactions and sorting,and then package the ordered transactions into blocks.The Orderer node is the core node of the Fabric system.However,the research found that the maintenance ability of the Fabric blockchain system in the face of the abnormality of the Orderer node still lacks an evolutionary mechanism at the software architecture level,that is,dynamic monitoring of system failures and automatic recovery of system operation mechanisms.There are also challenges.In view of the above problems,this article starts from the perspective of software architecture evolution methods,introduces the reflector of the Orderer node state,and discovers abnormal Orderer node states in time through the observer component;adopts a fault-tolerant software architecture method to increase the backup of the main Orderer node Nodes to realize the dynamic evolution of software architecture.An evolutionary algorithm of fault-tolerant software architecture is given,and the effectiveness of the algorithm is analyzed.This paper gives the experimental design and experimental results of the above software fault tolerance mechanism,and compares and analyzes the performance indicators such as the throughput and resource utilization of the fault tolerant software architecture.The design,implementation and testing of the prototype system based on the above algorithms are introduced.