Summary statistics are essential to analyse large datasets in various fields, including financial and medical research. Federated computations enhance statistical power by combining geo-distributed datasets while ensuring compliance with data protection regulations, privacy guarantees, and resilience against intrusions. We present Tides, a federated framework leveraging Trusted Execution Environments (TEEs) to defend against adversaries controlling up to f of the N datacenters. We present an instantiation of Tides using genomic (GWAS) statistics. We address TEE-specific attack vectors, including communication blocking and side-channel attacks. Tides follows three key steps: TEEs share statistical results through reliable broadcast and run randomized crash-tolerant binary consensus to identify available datasets; TEEs enforce differential privacy with ad hoc noise; and TEEs run memory-oblivious algorithms to compute the final summary statistics. We implemented Tides with Intel SGX enclaves and demonstrated its practicality with three datasets.
@inproceedings{segalini2025byzantine,title={Byzantine-Resilient Federated Computation of Differentially Private Summary Statistics},author={Segalini, Giulio and Fernandes, Maria and Decouchant, J{\'e}r{\'e}mie},booktitle={Middleware 2025 - Proceedings of the 26th ACM International Middleware Conference},publisher={Association for Computing Machinery},pages={72--85},year={2025},doi={10.1145/3721462.3770766},}