Towards Efficient and Verifiable SQL Queries with Trusted Hardware in Blockchain Systems

Blockchain allows mutually distrusting parties to make consistent and reliable agreement with- out a central authority. Over recent years, this innovative technology has been adopted as a trustworthy data storage and retrieval solution across many industries, such as finance, health- care, and supply chain management. However, most blockchain systems today are confined to fine-grained query types. How to support more general queries, such as full-fledged SQL queries, remains unexplored.

This project aims to investigate a new blockchain paradigm with trusted hardwares (e.g., Intel SGX) for efficient and verifiable SQL queries. The main idea is to maintain a succinct authen- ticated data structure on the chain, while complex query processing is outsourced to off-chain nodes equipped with trusted hardwares. The benefits are threefold. First, this approach can ease the computation burden of the smart contract, making query processing more efficient. Second, owing to the existence of trusted hardware, the integrity of query processing can be guaranteed at a much lower cost. SQL query results on multi-dimensional data can be efficiently veri- fied only with a cryptographic signature. Third, queries can be processed on off-chain nodes concurrently, improving system throughput.

There are several challenges to a realization of this paradigm. First, enabling users to verify general SQL queries on multi-dimensional data in blockchain systems is a fundamental issue. Second, with the constraint of memory size in SGX, queries should be processed with efficiency in terms of gas consumption, communication, storage, and computation. Third, query service should be resilient to the query request dynamics. To address these challenges, our research agenda includes: i) design of a new authenticated data structure and a set of commitment proto- cols to verify distributed data updates in hardware-assisted blockchain settings; ii) investigation of data verification algorithms for storage nodes to support low-cost off-chain storage verifica- tion; iii) development of a new block structure and efficient SQL query processing algorithms that are capable of block search and integrity verification; and iv) exploration of optimization techniques, such as off-chain storage layer sharding, batch proof verification, and on-SGX data replication, to further improve query concurrency and efficiency. Finally, we will collaborate with our industry partners (i.e., Peking University Shenzhen Hospital and Deshang Pharmaceu- tical Limited) for prototyping and testing in their blockchain service platform.

With our rich research experience in blockchain data management and verifiable computing, we expect the outcome of this project to accelerate the growth and adoption of blockchain technologies and decentralized services in the pertinent industries.