33

 

Automatic Compilation of Firewall and Intrusion Detection Rules for High-Speed Network Processing Engines—Reservoir Labs, Inc., 632 Broadway, Suite 803, New York, NY  10012-2614; 212-780-0527, http://www.reservoir.com

Dr. Kenneth Mackenzie, Principal Investigator, kenmac@reservoir.com

Dr. Richard Lethin, Business Official, lethin@reservior.com

DOE Grant No. DE-FG02-04ER84062

Amount:  $750,000

 

As networks move to 10 Gbps and beyond (including scientific networks within the Department of Energy) the need arises for high-speed security solutions capable of defending these networks from cyberattacks.  The current market supplies no solutions that operate at these speeds.  To address this problem, this project will develop a toolchain for automatically rendering signatures from an intrusion detection system (IDS) into high-speed signature detection engines that run on network processors.  Phase I demonstrated the feasbility of applying  mapping technology to the problem of rendering intrusion detection rulesets onto network processors.  A technique was identified for selecting the signature from the space of possible problem framings at compilation-time.  Phase II  will develop a prototype toolchain for rendering IDS signatures into high-speed signature detection engines that run on network processors.  The toolchain, which will target 10 Gbps on next-generation network processors and will be fully automatic, will be validated and verified.

 

Commercial Applications And Other Benefits as described by the awardee:  The high-speed signature-detection technology should find use in multiple network applications, especially intrusion detection, spam detection, and deep-inspection firewalls.  Because the system will run on network processors rather than fixed-function hardware, the product will have advantages in time-to-market, time-in-market, and price-performance.