ORQ: Complex Analytics on Private Data with Strong Security Guarantees

2025-09-12

AI Summary

Existing multi-party computation (MPC) systems suffer from quadratic computational overhead and poor efficiency in executing complex relational queries—particularly multi-way joins and aggregations—hindering scalable private data analysis. Method: We propose the first high-performance privacy-preserving analytics system tailored for complex relational queries in MPC. Our approach introduces: (1) a join protocol eliminating quadratic complexity; (2) dynamic online aggregation and generic oblivious operators to bound intermediate result sizes; and (3) a vectorized data-parallel execution engine with communication cost amortization. Contribution/Results: The system reduces end-to-end latency significantly over LAN/WAN deployments, supports datasets an order of magnitude larger than prior work, and—crucially—executes the full TPC-H benchmark (Scale Factor 10) for the first time within a pure MPC framework. This establishes a practical, efficient, and cryptographically strong foundation for large-scale collaborative analysis over private relational data.

Technology Category

• Machine Learning: Scalability of ML Systems
• Data Mining & Knowledge Management: Scalability, Parallel & Distributed Systems
• Search and Optimization: Distributed Search

Application Category

• Security and Privacy: Large-scale security measurements
• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

We present ORQ, a system that enables collaborative analysis of large private datasets using cryptographically secure multi-party computation (MPC). ORQ protects data against semi-honest or malicious parties and can efficiently evaluate relational queries with multi-way joins and aggregations that have been considered notoriously expensive under MPC. To do so, ORQ eliminates the quadratic cost of secure joins by leveraging the fact that, in practice, the structure of many real queries allows us to join records and apply the aggregations "on the fly" while keeping the result size bounded. On the system side, ORQ contributes generic oblivious operators, a data-parallel vectorized query engine, a communication layer that amortizes MPC network costs, and a dataflow API for expressing relational analytics -- all built from the ground up. We evaluate ORQ in LAN and WAN deployments on a diverse set of workloads, including complex queries with multiple joins and custom aggregations. When compared to state-of-the-art solutions, ORQ significantly reduces MPC execution times and can process one order of magnitude larger datasets. For our most challenging workload, the full TPC-H benchmark, we report results entirely under MPC with Scale Factor 10 -- a scale that had previously been achieved only with information leakage or the use of trusted third parties.

Problem

• Efficiently perform secure multi-party computation on private datasets
• Enable complex relational queries with joins and aggregations
• Protect data against semi-honest or malicious parties

Innovation

• Multi-party computation for private data analysis
• Eliminates quadratic cost of secure joins
• Generic oblivious operators and vectorized engine