Scalanytics
Harjot Gill, Dong Lin, Xianglong Han, Cam Nguyen, Tanveer Gill, Boon Thau Loo · 2013
This paper presents SCALANYTICS, a declarative platform that supports high-performance application layer analysis of network traffic. SCALANYTICS uses (1) stateful network packet processing techniques for extracting application-layer data from network packets, (2) a declarative rule-based language called ANALOG for compactly specifying analysis pipelines from reusable modules, and (3) a task-stealing architecture for processing network packets at high throughput within these pipelines, by leveraging multi-core processing capabilities in a load-balanced manner without the need for explicit performance profiling. We have developed a prototype of SCALANYTICS that enhances a declarative networking engine with support for ANALOG and various stateful components, integrated with a parallel task-stealing execution model. We evaluate our SCALANYTICS prototype on a wide range of pipelines for analyzing SMTP and SIP traffic, and for detecting malicious traffic flows. Our evaluation on a 16-core machine demonstrate that SCALANYTICS achieves up to 11.4× improvement in throughput compared with the best uniprocessor implementation. Moreover, SCALANYTICS outperforms the Bro intrusion detection system by an order of magnitude when used for analyzing SMTP traffic.