Abstract

The packet classification technique play an important role for many internet advanced services that require the capability to distinguish traffic in different flows, such as routers and security services like firewalls and packet filters. This paper provide a brief introduction for packet classification process. Beside, the categorization of the algorithms that classify packets to different flow. Also, this paper give a brief description of two homogenous algorithms (algorithms that belongs to same category), these algorithms are Tuple Space Search (TSS) and Recursive Flow Classification (RFC). Furthermore, this paper provides a comparative evaluation of these two algorithms (TSS & RFC). The evaluation of these techniques are done under Linux REDHAT 5.0 platform in order to analyze and compare their performance between each other. In particular, the comparison focuses on achieving a good classification time in classification process and, low memory consumption rate.

Highlights

  • Packet classifiers are essential components of many network utilities, including routers and security services like firewalls and packet filters

  • Recursive Flow Classification (RFC) and Tuple Space Search (TSS) algorithms classified the data that has been captured in order to evaluate the performance for both algorithms

  • This experiment aims to giving a homogeneous comparison between two algorithms that categorized under same category, to determine the advantages and limitations for each of the algorithms; and this evaluation considering two kinds of metric that will describe later

Read more

Summary

Introduction

Packet classifiers are essential components of many network utilities, including routers and security services like firewalls and packet filters. Most of packet classification processes happens in the internet [1]. An IP router may choose to perform special processing on incoming packets as described earlier. Examples of special processing include filtering packets for security reasons, delivering packets to destination hosts, and treating high priority packets preferentially. Such special processing requires that the router classify incoming packets into one of several flows. To illustrate the variety of classifiers, this study will consider some examples of packet classification mechanism that can be used by an ISP (Internet Server Provider) to provide different services.

Existing Packet Classification Algorithms
Decomposition-based
Experiment Results
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call