Secure parallel computation on national scale volumes of data

Authors: 

Sahar Mazloom and Phi Hung Le, George Mason University; Samuel Ranellucci, Unbound Tech; S. Dov Gordon, George Mason University

Abstract: 

We revisit the problem of performing secure computation of graph-parallel algorithms, focusing on the applications of securely outsourcing matrix factorization, and histograms. Leveraging recent results in low-communication secure multi-party computation, and a security relaxation that allows the computation servers to learn some differentially private leakage about user inputs, we construct a new protocol that reduces overall runtime by 320X, reduces the number of AES calls by 750X , and reduces the total communication by 200X . Our system can securely compute histograms over 300 million items in about 4 minutes, and it can perform sparse matrix factorization, which is commonly used in recommendation systems, on 20 million records in about 6 minutes. Furthermore, in contrast to prior work, our system is secure against a malicious adversary that corrupts one of the computing servers.

Open Access Media

USENIX is committed to Open Access to the research presented at our events. Papers and proceedings are freely available to everyone once the event begins. Any video, audio, and/or slides that are posted after the event are also free and open to everyone. Support USENIX and our commitment to Open Access.

BibTeX
@inproceedings {247676,
author = {Sahar Mazloom and Phi Hung Le and Samuel Ranellucci and S. Dov Gordon},
title = {Secure parallel computation on national scale volumes of data},
booktitle = {29th USENIX Security Symposium (USENIX Security 20)},
year = {2020},
isbn = {978-1-939133-17-5},
pages = {2487--2504},
url = {https://www.usenix.org/conference/usenixsecurity20/presentation/mazloom},
publisher = {USENIX Association},
month = aug
}

Presentation Video