ACHIEVING 100 GB/S URL FILTERING WITH COTS MULTI-CORE SYSTEMS

Authors

  • Surachai Chitpinityon Kasetsart University
  • Surasak Sanguanpong Kasetsart University
  • Supaporn Erjongmanee Kasetsart University
  • Kasom Koht-Arsa Department of Computer Engineering Faculty of Engineering

DOI:

https://doi.org/10.7903/ijecs.1483

Keywords:

URL Filtering, Web Filtering, Session Hijacking, 100 GbE, COTS, AVL

Abstract

URL filtering is an essential tool used by Internet Service Providers (ISPs) and organizations to restrain clients from accessing non-secured or illegal web content. Designing a URL filtering method that achieves a high bit rate of 100 Gb/s and beyond for international ISPs is a challenging task. High-performance URL filtering with multi-gigabit rate capacity requires a fast URL matching algorithm and an enhanced packet processing technique. In this paper, we tackle these challenges by design and development of a software-based URL filtering system to support 100 Gb/s bandwidth. Our aim is to build a system that runs on a single commercial off-the-shelf (COTS) server with multi-core CPUs. We propose a compact URL representation using AVL tree and a multi-core/multi-thread filtering technique with session hijacking and fast packet processing framework. Performance measurements results show successful URL filtering operating at 100 Gb/s in a real network testbed.

To cite this document: Surachai Chitpinityon, Surasak Sanguanpong, Supaporn Erjongmanee, and Kasom Koht-Arsa, "Achieving 100 GB/S URL filtering with cots multi-core systems", International Journal of Electronic Commerce Studies, Vol.8, No.1, pp.77-96, 2017.

Permanent link to this document:
http://dx.doi.org/10.7903/ijecs.1483

 

Author Biographies

Surachai Chitpinityon, Kasetsart University

Department of Computer Engineering
Faculty of Engineering

Surasak Sanguanpong, Kasetsart University

Department of Computer Engineering
Faculty of Engineering

Supaporn Erjongmanee, Kasetsart University

Department of Computer Engineering
Faculty of Engineering

Kasom Koht-Arsa, Department of Computer Engineering Faculty of Engineering

Department of Computer Engineering
Faculty of Engineering

Downloads

Published

2017-07-30

Issue

Section

Systems/websites development and application issues