Shadow-Chain: A Decentralized Storage System for Log Data

Abstract

Better use of personal health and clinical data from different stakeholders is essential and highly significant. However, the sharing of such data could easily raise concerns over preserving personal privacy, particularly when the risk of disclosure is uncertain. To tackle this issue, shadow- chain, a decentralized storage system for log data based on blockchain, is proposed for realizing data sharing between parties that do not trust each other. In the system, a behavior monitor is designed to intercept the data reading activities. Through an asynchronous read-write mechanism, the monitor can minimize operational impact on the source's storage system. Meanwhile, the homomorphic encryption scheme and the distributed log collection system are adopted to guarantee the validity and reliability of the log data. Furthermore, the hybrid storage strategy is developed to reduce the storage overhead of the log data. In the strategy, only the hash of log data is stored in the blockchain of the shadow-chain, while the data itself and the associated index information are placed in common storage media. Experiments have been conducted to estimate the performance of our shadow-chain. The results show that the impact of shadow-chain on the source storage system can be controlled within 5 percent, and 75 percent log storage overhead can be reduced through hybrid storage strategy.