Distributed & Scalable Oblivious Sorting and Shuffling
Nicholas Ngai, Ioannis Demertzis, Javad Ghareh Chamani, Dimitrios Papadopoulos
IEEE Symposium on Security and Privacy 2024 · Day 3 · Continental Ballroom 6
In an era where data privacy is paramount, traditional encryption alone often proves insufficient to protect sensitive information from sophisticated attacks. This talk, presented by Nicholas Ngai and his colleagues from UC Berkeley, UC Santa Cruz, and HK, delves into the critical challenge of ensuring data confidentiality not just at rest or in transit, but also during computation. The core problem addressed is **side-channel leakage**, where adversaries can infer sensitive data by observing patterns in memory access, control flow, or network requests, even when the data itself is encrypted or processed within secure environments like hardware enclaves.
AI review
This work shatters the scalability barrier for oblivious sorting and shuffling, a critical bottleneck for real-world privacy-preserving computation. The dBucket sort, with its O(N) network communication and novel optimizations, delivers unprecedented performance and capacity for large datasets. This isn't just theory; it directly enables practical, high-throughput privacy for applications from contact discovery to LLMs.