An Efficient Data Query Method of Blockchain Based on Index

In many blockchain applications, apart from storing the original data, it is also necessary store the data operation records in the blockchain for audit or security. Yet, the traditional blockchain, generally cannot meet the user's demand for data query. In order to solve this problem, we propose a novel block structure, the block body into an index layer and a data layer. On this basis, we further develop two in the index layer to aggregate the data and its operation records for efficient query. Specifically, we first build an index for the original data in blockchain by desigining the Abstract-Trie, which has efficient and stable query features. Secondly, we design another index called Operation-Record List to link the operation records to the corresponding original data, and store the first node of each list in the Abstract-Trie, so as to achieve the fast acquisition of the operation records of the original data. Simulation results based on Hyperledger Fabric verify the effectiveness of our optimized block structure in querying of the original data and its operation records in the blockchain.

I. Introduction

Owing to the success of tranditional blockchain such as Bitcoin[1], Ethereum[2] and Hyperledger Fabric[3], blockchain technology has attracted much attention from both academia and industry. With the wide application of blockchain technology in distributed systems, how to efficiently obtain data in blockchain has become a new research hotspot. As shown in Fig. 1, for the traditional blockchain, data is stored in each node of the network in the form of blocks. A block consists of a block header and a block body, where the block header mainly stores the basic information, such as the hash value of the previous block, and the block body stores the actual data. It is worth mentioning that the white paper of blockchain technology does not strictly define the organization form of the data, which suggests that we should traverse all the data in each block in order to acquire the required data.