CloudDPI: Cloud-Based Privacy-Preserving Deep Packet Inspection Via Reversible Sketch.

Jie Li,Jinshu Su,Xiaofeng Wang,Hao Sun,Shuhui Chen

DOI: https://doi.org/10.1007/978-3-319-69471-9_9

2017-01-01

Abstract:Hardware-based middleboxes are ubiquitous in computer networks, which usually incur high deployment and management expenses. A recently arsing trend aims to address those problems by outsourcing the functions of traditional hardware-based middleboxes to high volume servers in a cloud. This technology is promising but still faces a few challenges. First, the widely adopted data encryption techniques contradict with payload inspection needs of some middleboxes such as DPI and IDS devices. Second, the inspection rules of middleboxes may be commercial properties, thus the middlebox providers want to keep their rules confidential under third-party cloud environments, and this creates hindrances for the cloud to perform outsourced middlebox functions. Third, performance of the outsourced middlebox is an inevitable issue that needs deliberate consideration. In this paper, we propose a cloud-based DPI middlebox implementation which performs payload inspection over encrypted traffic while preserving the privacy of both communication data and inspection rules. Our design employs a modified reversible sketch structure which is used for efficient error-free membership testing, and we utilize unkeyed one-way hash functions instead of complex cryptographic protocols to achieve the privacy preservation requirements. CloudDPI supports a wide range of real-world inspection rules, we conduct evaluations on ClamAV rule set and the experiment results demonstrate the effectiveness of our proposal.