ObliviSync: Practical Oblivious File Backup and Synchronization
May 31, 2016 ยท Declared Dead ยท ๐ Network and Distributed System Security Symposium
"No code URL or promise found in abstract"
Evidence collected by the PWNC Scanner
Authors
Adam J. Aviv, Seung Geol Choi, Travis Mayberry, Daniel S. Roche
arXiv ID
1605.09779
Category
cs.CR: Cryptography & Security
Citations
13
Venue
Network and Distributed System Security Symposium
Last Checked
3 months ago
Abstract
Oblivious RAM (ORAM) protocols are powerful techniques that hide a client's data as well as access patterns from untrusted service providers. We present an oblivious cloud storage system, ObliviSync, that specifically targets one of the most widely-used personal cloud storage paradigms: synchronization and backup services, popular examples of which are Dropbox, iCloud Drive, and Google Drive. This setting provides a unique opportunity because the above privacy properties can be achieved with a simpler form of ORAM called write-only ORAM, which allows for dramatically increased efficiency compared to related work. Our solution is asymptotically optimal and practically efficient, with a small constant overhead of approximately 4x compared with non-private file storage, depending only on the total data size and parameters chosen according to the usage rate, and not on the number or size of individual files. Our construction also offers protection against timing-channel attacks, which has not been previously considered in ORAM protocols. We built and evaluated a full implementation of ObliviSync that supports multiple simultaneous read-only clients and a single concurrent read/write client whose edits automatically and seamlessly propagate to the readers. We show that our system functions under high work loads, with realistic file size distributions, and with small additional latency (as compared to a baseline encrypted file system) when paired with Dropbox as the synchronization service.
Community Contributions
Found the code? Know the venue? Think something is wrong? Let us know!
๐ Similar Papers
In the same crypt โ Cryptography & Security
R.I.P.
๐ป
Ghosted
R.I.P.
๐ป
Ghosted
Membership Inference Attacks against Machine Learning Models
R.I.P.
๐ป
Ghosted
The Limitations of Deep Learning in Adversarial Settings
R.I.P.
๐ป
Ghosted
Practical Black-Box Attacks against Machine Learning
R.I.P.
๐ป
Ghosted
Distillation as a Defense to Adversarial Perturbations against Deep Neural Networks
R.I.P.
๐ป
Ghosted
Extracting Training Data from Large Language Models
Died the same way โ ๐ป Ghosted
R.I.P.
๐ป
Ghosted
Language Models are Few-Shot Learners
R.I.P.
๐ป
Ghosted
PyTorch: An Imperative Style, High-Performance Deep Learning Library
R.I.P.
๐ป
Ghosted
XGBoost: A Scalable Tree Boosting System
R.I.P.
๐ป
Ghosted