Privacy-Enhanced Database Synthesis for Benchmark Publishing (Technical Report)

Abstract

Benchmarking is crucial for evaluating a DBMS, yet existing benchmarks often fail to reflect the varied nature of user workloads. As a result, there is increasing momentum toward creating databases that incorporate real-world user data to more accurately mirror business environments. However, privacy concerns deter users from directly sharing their data, underscoring the importance of creating synthesized databases for benchmarking that also prioritize privacy protection. Differential privacy (DP)-based data synthesis has become a key method for safeguarding privacy when sharing data, but the focus has largely been on minimizing errors in aggregate queries or downstream ML tasks, with less attention given to benchmarking factors like query runtime performance. This paper delves into differentially private database synthesis specifically for benchmark publishing scenarios, aiming to produce a synthetic database whose benchmarking factors closely resemble those of the original data. Introducing \textit{PrivBench}, an innovative synthesis framework based on sum-product networks (SPNs), we support the synthesis of high-quality benchmark databases that maintain fidelity in both data distribution and query runtime performance while preserving privacy. We validate that PrivBench can ensure database-level DP even when generating multi-relation databases with complex reference relationships. Our extensive experiments show that PrivBench efficiently synthesizes data that maintains privacy and excels in both data distribution similarity and query runtime similarity.

Technical report for our VLDB 2025 paper. Please cite the original publication: https://dl.acm.org/doi/abs/10.14778/3705829.3705855

References in corpus (13)

• Differentially Private Generative Adversarial Network
• Differential Privacy: An Economic Method for Choosing Epsilon
• Learned Cardinalities: Estimating Correlated Joins with Deep Learning
• Computing Local Sensitivities of Counting Queries with Joins
• Graphical-model based estimation and inference for differential privacy
• Secure Shapley Value for Cross-Silo Federated Learning (Technical Report)
• GS-WGAN: A Gradient-Sanitized Approach for Learning Differentially Private Generators
• New Oracle-Efficient Algorithms for Private Synthetic Data Release
• G-PATE: Scalable Differentially Private Data Generator via Private Aggregation of Teacher Discriminators
• Kamino: Constraint-Aware Differentially Private Data Synthesis
• AIM: An Adaptive and Iterative Mechanism for Differentially Private Synthetic Data
• CryptoSPN: Privacy-preserving Sum-Product Network Inference
• PrivLava: Synthesizing Relational Data with Foreign Keys under Differential Privacy