Data storage method and system, apparatus, and storage medium

Abstract

The invention provides a data storage method and a system, an apparatus and a storage medium. The method comprises the following steps: a global index table corresponding to a Merkel tree of each block is stored locally at each node of a block chain network; and a Merkel tree of each block is stored locally. Where the version number of the version is the block height, and the global index table isused to query data according to the block height. By configuring the corresponding global index table for the Merkel tree and taking the block height as the version number, the invention provides a data inquiry way for the system to inquire the global index table directly through the block height and complete the inquiry through one reading operation, thereby greatly improving the data reading performance of the system. And further optimizes the robustness of the system in the event of transaction congestion or node crash.

Description

technical field [0001] The present application relates to the technical field of block chain, in particular to a data storage method, system, equipment and storage medium. Background technique [0002] In the current data reading and writing solutions of the blockchain system, the structure of the Merkle tree is usually adopted. Kerr prefix tree (Merkle Patricia Tree, referred to as MPT) for data reading and writing, and so on. [0003] The shortcomings of the current data storage scheme using the Merkle tree structure are: [0004] The stored data limits the read performance of the system, and querying the data of a transaction requires multiple read operations. For example, for a 20-layer Merkle tree, querying the data of a leaf node requires 20 read operations to complete, resulting in the efficiency of data query is only 1 / 20 of the query efficiency of ordinary databases, which can be completed per second A system with 100,000 read operations can only read data of 5,0...

AI Technical Summary

Problems solved by technology

[0004] The stored data limits the read performance of the system, and querying the data of a transaction requires multiple read operations

For example, for a 20-layer Merkle tree, querying the data of a leaf node requires 20 read operations to complete, resulting in the efficiency of data query is only 1 / 20 of the query efficiency of ordinary databases, which can be completed per second A system with 100,000 read operations can only read data of 5,000 transactions per second

[0005] Furthermore, the existing scheme needs to complete the block consensus after writing the Merkle tree data in the local node database, which leads to transaction congestion when the number of transactions increases sharply

[0006] In addition, if the existing scheme crashes when writing Merkle tree data in the local database of the node, it will result in failure to generate blocks

Images