Opaque: An Oblivious And Encrypted Distributed Analytics Platform
PROCEEDINGS OF NSDI '17: 14TH USENIX SYMPOSIUM ON NETWORKED SYSTEMS DESIGN AND IMPLEMENTATION(2017)
Abstract
Many systems run rich analytics on sensitive data in the cloud, but are prone to data breaches. Hardware enclaves promise data confidentiality and secure execution of arbitrary computation, yet still suffer from access pattern leakage. We propose Opaque, a distributed data analytics platform supporting a wide range of queries while providing strong security guarantees. Opaque introduces new distributed oblivious relational operators that hide access patterns, and new query planning techniques to optimize these new operators. Opaque is implemented on Spark SQL with few changes to the underlying system. Opaque provides data encryption, authentication and computation verification with a performance ranging from 52% faster to 3.3x slower as compared to vanilla Spark SQL; obliviousness comes with a 1.6-46x overhead. Opaque provides an improvement of three orders of magnitude over state-of-the-art oblivious protocols, and our query optimization techniques improve performance by 2-5x.
MoreTranslated text
Key words
analytics platform
AI Read Science
Must-Reading Tree
Example
![](https://originalfileserver.aminer.cn/sys/aminer/pubs/mrt_preview.jpeg)
Generate MRT to find the research sequence of this paper
Chat Paper
Summary is being generated by the instructions you defined