Research And Development Of High Real-time Transaction System Based On Hyperledger Fabric

Currently,data storage methods mainly include centralized storage and distributed storage.Data stored in a centralized manner is vulnerable to attacks and security cannot be guaranteed;when the server storing the data crashes,a large amount of data may be lost and the entire system is unusable.The blockchain technology uses distributed data storage can effectively prevent data from being tampered with mali-ciously and ensure data security.However,most of the current work focuses on using blockchain technology to implement applications,ignoring the problems of blockchain technology itself,especially the real-time issue.At present,an important reason why blockchain technology has not achieved large-scale applications is that the system based on blockchain technology lacks real-time performance and cannot meet real application requirements.Therefore,it is urgent to enhance the real-time of the blockchain-based system.This thesis develops and researches a high real-time transaction system based on Hyperledger Fabric.In order to improve the real-time performance of the system based on Hyperledger Fabric,this thesis mainly researches from two aspects:condi-tional queries and data storage.Finally,a prototype system was implemented on this basis.In the realization of high real-time conditional queries based on Hyperledger Fab-ric,this thesis mainly considers two aspects:firstly,how to improve the efficiency of conditional queries and reduce the waiting time of conditional queries;secondly,on the basis of enhancing the efficiency of conditional queries,to reduce data redun-dancy.To this end,this thesis proposes two methods to execute conditional queries on Hyperledger Fabric:CCKM and AIM.All methods take into account the impact of the total amount of data and data distribution on the efficiency of conditional queries and there is a progressive relationship between them.The AIM method has the best performance,it not only improves the response time of conditional queries,but also reduces data redundancy.In this thesis,a large number of experiments are performed on the real data set for verification,analysis and summaries are also given.In the realization of high real-time data storage on Hyperledger Fabric,this thesis studies how to increase the speed of storing transaction data in the ledger.In response to this problem,a method CID for storing the data to be verified in the block into the cache is proposed,thereby reducing the time to deserialize the block.In order to further improve the efficiency of data storage,this thesis designs the method CMM,which uses MemoryDB to implement the state database on the basis of CID,thereby improving the access speed of the state database.Finally,after a lot of experiments,it is proved that both CID and CMM improve the efficiency of data storage.In the realization of high real-time transaction system based on Hyperledger Fab-ric,this thesis designs an smart contract and use Go to realize the main functions of the system.The system has two user roles:ordinary user and administrator user,in which administrator users can manage the whole system,and ordinary users can log in to the system and initiate transactions,including executing queries and storing transaction data.The research content of this thesis greatly improves the real-time performance of the transaction system based on Hyperledger Fabric,and provides a new idea for realizing high real-time processing of transactions based on blockchain.