QueryShield: cryptographically secure analytics in the cloud

Files
3626246.3654749.pdf(1.28 MB)
Published version
Date
2024-06-09
Authors
Seow, Ethan
Baum, Eli
Buxbaum, Sam
Faisal, Muhammad Sajid
Liagouris, John
Kalavri, Vasiliki
Varia, Mayank
Version
OA Version
Citation
Ethan Seow, Yan Tong, Eli Baum, Sam Buxbaum, Muhammad Faisal, John Liagouris, Vasiliki Kalavri, and Mayank Varia. 2024. QueryShield: Cryptographically Secure Analytics in the Cloud. In Companion of the 2024 International Conference on Management of Data (SIGMOD/PODS '24). Association for Computing Machinery, New York, NY, USA, 436–439. https://doi.org/10.1145/3626246.3654749
Abstract
We present a demonstration of QueryShield, a service for streamlined, cryptographically secure data analytics in the cloud. With QueryShield, data analysts can advertise analysis descriptions to data owners, who may agree to participate in a computation for profit or for the greater good, provided that their data remain private. QueryShield supports relational and time series analytics with provable data privacy guarantees using secure multi-party computation (MPC). At the same time, it makes MPC accessible to non-expert users by offering a familiar web interface and fully-automated orchestration of cryptographic computations. We devise three demonstration scenarios for conference attendees: (i) an interactive survey of private employment information to estimate the industry-academia wage gap in the data management community, (ii) a relational analysis that identifies credit score anomalies in sensitive customer data from multiple credit agencies, and (iii) a medical use case that assesses the effectiveness of insulin dose frequency in a patient cohort.
Description
License
© 2024 Copyright held by the owner/author(s). This work is licensed under a Creative Commons Attribution International 4.0 License. This article has been published under a Read & Publish Transformative Open Access (OA) Agreement with ACM.