Shunting: a hardware/software architecture for flexible, high-performance network intrusion prevention

  • Authors:
  • Jose M. Gonzalez;Vern Paxson;Nicholas Weaver

  • Affiliations:
  • International Computer Science Institute, Berkeley, CA;International Computer Science Institute, Berkeley, CA;International Computer Science Institute, Berkeley, CA

  • Venue:
  • Proceedings of the 14th ACM conference on Computer and communications security
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

Stateful, in-depth, inline traffic analysis for intrusion detection and prevention is growing increasingly more difficult as the data rates of modern networks rise. Yet it remains the case that in many environments, much of the traffic comprising a high-volume stream can, after some initial analysis, be qualified as of "likely uninteresting." We present a combined hardware/software architecture, Shunting, that provides a lightweight mechanism for an intrusion prevention system (IPS) to take advantage of the "heavy-tailed" nature of network traffic to offload work from software to hardware. The primary innovation of Shunting is the introduction of a simple in-line hardware element that caches rules for IP addresses and connection 5-tuples, as well as fixed rules for IP/TCP flags. The caches, using a highest-priority match, yield a per-packet decision: forward the packet; drop it; or divert it through the IPS. By manipulating cache entries, the IPS can specify what traffic it no longer wishes to examine, including directly blocking malicious sources or cutting through portions of a single flow once the it has had an opportunity to "vet" them, all on a fine-grained basis. We have implemented a prototype Shunt hardware design using the NetFPGA 2 platform, capable of Gigabit Ethernet operation. In addition, we have adapted the Bro intrusion detection system to utilize the Shunt framework to offload less-interesting traffic. We evaluate the effectiveness of the resulting system using traces from three sites, finding that the IDS can use this mechanism to offload 55%-90% of the traffic, as well as gaining intrusion prevention functionality.