Zeph: Cryptographic Enforcement of End-to-End Data Privacy
Talkat 15th USENIX Symposium on Operating Systems Design and Implementation (OSDI '21) · 2021-07-14..07-16·
summary
Zeph enforces user-defined privacy by encrypting data at sources and issuing cryptographic transformation tokens from a separate privacy plane so services can only decrypt privacy-compliant views. It implements additive homomorphic secret sharing to compute masking, aggregation and calibrated differential privacy noise on encrypted streams, integrates with Apache Kafka (Java and Rust prototype), and runs end-to-end across AWS with 300–1,200 producers/controllers, showing sub-second overhead at 1,200 participants versus plaintext.
AI-written summary · about
chapters (9)
- 0:00Data privacy challenges
- 1:59Introduction to Ceph
- 3:20Privacy transformation goals
- 4:24Core architecture design
- 6:24Privacy control mechanics
- 8:10System-level challenges
- 9:20Cryptographic foundations
- 12:22Advanced security features
- 13:31Evaluation and conclusion
related talks
38mPrivacy Enhancing Cryptography at NIST
19mInnovations in Privacy Technologies: The Keys To Decentralized Networks
39mData Privacy Engineering to power Responsible Enterprise AI
13mSurvey of Privacy Solutions for Minimizing MEV
40mKeeping data private in real-time pipelines
46mCan we have privacy and compliance at the same time?
24mSafeguarding Sensitive Data Access At Scale with Privacy-Enhancing Technologies
26mZether Towards Privacy in a Smart Contract World