An Abstract Model for Intrusion Detection on Multi-Core Platform

Author

Zhuang, Zhuojun ; Luo, Yuan ; Li, Minglu ; Weng, Chuliang

Author_Institution

Dept. of Comput. Sci. & Eng., Shanghai Jiao Tong Univ., Shanghai

fYear

2008

fDate

20-22 Aug. 2008

Firstpage

202

Lastpage

208

Abstract

The processing speed of conventional network-based intrusion detection systems (NIDSs) is incompetent as to the rapid increase in network link speed. This problem imposes an emerging need for new detection technologies. In this paper, we introduce a multi-core technique which opens up another way for fast intrusion detection by proper workload partitioning and parallel detection on high-speed link. We present an abstract model of NIDS on multi-core platform and discuss the optimal solution when minimum memory occupation achieves. A preliminary example shows the model configuration and presents a first experimental evaluation.

Keywords

multiprocessing systems; security of data; high-speed link; intrusion detection systems; multicore platform; multicore technique; network link speed; parallel detection; workload partitioning; Central Processing Unit; Computer science; Concurrent computing; High-speed networks; Intrusion detection; Measurement; Multicore processing; Protection; Telecommunication traffic; Throughput;

fLanguage

English

Publisher

ieee

Conference_Titel

ChinaGrid Annual Conference, 2008. ChinaGrid '08. The Third

Conference_Location

Dunhuang, Gansu

Print_ISBN

978-0-7695-3306-3

Type

conf

DOI

10.1109/ChinaGrid.2008.27

Filename

4624490