Privacy-Preserving Federated Query Processing Across Distributed Cloud Data Platforms
International Journal of Engineering, Science and Information Technology · 25.04.2026 · 6 (2) · 117-122
DOI: 10.52088/ijesty.v6i2.1844 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Engineering, Science and Information Technology, online erschienen am 25.04.2026. Verfasst von Shankar Das Boddu. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
The rapid adoption of multi-cloud data platforms has enabled organizations to perform large-scale distributed analytics while complying with regional data governance requirements. However, existing federated query processing frameworks often prioritize performance and interoperability without providing rigorous privacy guarantees or enforcing regulatory compliance across multiple jurisdictions. This study presents PrivFed, a privacy-preserving federated query processing framework designed for distributed cloud data platforms operating under heterogeneous legal and regulatory constraints. PrivFed integrates differential privacy (DP), secure aggregation (SA), and a region-aware multi-objective query optimizer to support compliant, low-latency analytics over geographically partitioned datasets without exposing sensitive information. The proposed framework formalizes the federated query compliance problem by jointly optimizing query execution cost, privacy preservation, and data residency requirements. Formal analysis establishes -DP guarantees under adaptive composition and proves the security of the aggregation protocol in the semi-honest adversarial model. A comprehensive prototype was implemented across AWS Redshift, Azure Synapse, and Google BigQuery spanning three regulatory regions to evaluate scalability, efficiency, and privacy performance. Experimental results demonstrate that PrivFed achieves a median query latency only 1.4× higher than conventional non-private federated query systems, substantially outperforming homomorphic encryption-based approaches that incur 18–340× latency overhead. Furthermore, privacy-aware predicate pushdown reduces inter-region data transfer by 62%, while maintaining a cumulative privacy budget of ? ? 1.0 across 10,000 simulated adaptive queries. Comparative evaluation against Presto, Trino, and BigQuery Omni indicates that PrivFed is the only framework capable of simultaneously satisfying three critical objectives: strict data residency compliance, mathematically provable privacy protection, and practical query execution with less than 2× performance overhead. These findings demonstrate that PrivFed provides a practical and scalable foundation for secure federated analytics in modern multi-cloud environments.
Zitieren
APA
Boddu, S. D. (2026). Privacy-Preserving Federated Query Processing Across Distributed Cloud Data Platforms. International Journal of Engineering, Science and Information Technology, 6(2), 117-122. https://doi.org/10.52088/ijesty.v6i2.1844
Vancouver
Boddu SD. Privacy-Preserving Federated Query Processing Across Distributed Cloud Data Platforms. International Journal of Engineering, Science and Information Technology. 2026;6(2):117-122. doi:10.52088/ijesty.v6i2.1844
BibTeX
@article{Boddu2026,
author = {Boddu, Shankar Das},
title = {Privacy-Preserving Federated Query Processing Across Distributed Cloud Data Platforms},
journal = {International Journal of Engineering, Science and Information Technology},
year = {2026},
volume = {6},
number = {2},
pages = {117--122},
doi = {10.52088/ijesty.v6i2.1844}
}
EUVIMED