Please use this identifier to cite or link to this item:
http://dspace.cityu.edu.hk/handle/2031/547
Title: | Fast packet classification algorithms I |
Authors: | Leung, Kai Sum |
Department: | Department of Computer Engineering and Information Technology |
Issue Date: | 2003 |
Supervisor: | Dr. Pao Derek C W. Assessor: Dr. Wong K W |
Abstract: | IP routers use packet classification to support advanced functions such as quality of services (QoS), virtual private networks (VPN), firewalls, and access control. The rules used for classifying packets are called filters. A filter is made up of multiple header fields, e.g. the source IP address, destination IP address, protocol type, and/or port number. The packet classification problem is to determine the best matching filter (the highest priority filter that matches the packet's header fields) in a filter database. In real life, a filter database may have over 10,000 filters. The challenge of the problem is the tight real-time requirement. To match the speed of a gigabit backbone router, the packet classification algorithm is required to sustain a throughput of several millions classifications per second. |
Appears in Collections: | Computer Engineering & Information Technology - Undergraduate Final Year Projects |
Files in This Item:
File | Size | Format | |
---|---|---|---|
fulltext.html | 164 B | HTML | View/Open |
Items in Digital CityU Collections are protected by copyright, with all rights reserved, unless otherwise indicated.