| At present,blockchain technology is increasingly concerned by governments,organizations and enterprises all over the world.It is widely used in credit investigation,financing,bonds and other fields.As a blockchain project with the largest number of users and projects,Ethereum has attracted more and more attention.As an early product of blockchain,Ethereum uses Merkle Patricia Tree(MPT)supported by Trie structure as the data storage structure.In recent years,the public has gradually deepened the research and application of blockchain data,and expects to optimize the blockchain to improve its performance.Based on this background,through the analysis and measurement of Ethereum structure and network data,this thesis concludes that Merkle Patricia Tree has some shortcomings such as low efficiency and complex operation in terms of network throughput,transaction confirmation time and data storage.Two different Ethereum data storage structures are designed by using Merkle Mountain Range(MMR),and its performance is verified by experiments.The main research contents of this thesis are as follows:(1)The technical principle of Ethereum is deeply analyzed and summarized.The Ethereum structure is decomposed into eight layer models to more clearly show the design structure of Ethereum.Based on this,the existing problems of Ethereum are positioned and analyzed.At the same time,the knowledge and technology of Merkle Mountain Range,Merkle Patricia Tree and State Machine Replication are analyzed and summarized.These technologies are the theoretical basis and technical support for the improvement and optimization of Ethereum data storage structure in this thesis.(2)Use the rust language to write the Ethereum network data acquisition tool,synchronize the Ethereum main network data,analyze and measure the data,discuss the analysis results with the Ethereum structure,draw the problems existing in Ethereum in data processing and resource occupation,and put forward solutions.(3)To solve the above problems,this thesis proposes an improved structure for Ethereum Merkle Patricia Tree: adapt Ethereum based on Merkle Mountain Range,design a hierarchical MMR structure supported by Merkle Mountain Range,and replace leveldb with rocksdb database with stronger performance and more suitable for this structure,By building a new database table,we can improve the performance and provide complex functions such as real-time tracking of account status.Based on the hierarchical MMR structure,the SMR-MMR structure with more emphasis on node data synchronization performance is designed by introducing State Machine Replication.Through experimental analysis,it is verified that the two structures proposed in this thesis have higher performance than the original structure in data storage and network node synchronization. |
PDF Full Text Request