Practical QoS network system with fault tolerance

  • Authors:
  • S. S. Lee;S. Das;H. Yu;K. Yamada;G. Pau;M. Gerla

  • Affiliations:
  • Computer Science Department, University of California, Los Angeles, CA 90095-1596, USA;Computer Science Department, University of California, Los Angeles, CA 90095-1596, USA;Computer Science Department, University of California, Los Angeles, CA 90095-1596, USA;Computer Science Department, University of California, Los Angeles, CA 90095-1596, USA;Computer Science Department, University of California, Los Angeles, CA 90095-1596, USA;Computer Science Department, University of California, Los Angeles, CA 90095-1596, USA

  • Venue:
  • Computer Communications
  • Year:
  • 2003

Quantified Score

Hi-index 0.24

Visualization

Abstract

In this paper, we present an 'emulation environment' for the design and planning of QoS-enabled IP networks. Specifically, QoS support for real-time traffic in future IP networks require 'fault tolerance' for mission-critical applications such as tele-medicine. Based on this premise, our goal is to design the QoS network system for the traffic management so that QoS requirements are satisfied. Routing is a key component of this architecture, in which it enables the efficient use of the provisioned bandwidth. Our approach is to make routers in the network capable of finding QoS paths in a distributed manner in response to changing traffic demands. As an extension of QoS routing, each router is able to compute 'multiple QoS paths' to the same destination. The multiple QoS paths are utilized in parallel making the entire network system less prone to network failures. In computing multiple QoS paths, practical issues such as computation latency and resource requirement must be considered. With the use of multiple QoS paths, our emulation environment permits us to design more reliable QoS network systems and to test various applications for feasibility before deployment. This paper presents the multiple QoS algorithms with notable enhancements and the system architecture for fault tolerance. The relevant experiment results have also been presented.