Comparative Analysis of SMP Click Scheduling Techniques

  • Giorgio Calarco
  • Carla Raffaelli
  • Giovanni Schembra
  • Giovanni Tusa
Conference paper

DOI: 10.1007/978-3-540-30573-6_29

Part of the Lecture Notes in Computer Science book series (LNCS, volume 3375)
Cite this paper as:
Calarco G., Raffaelli C., Schembra G., Tusa G. (2005) Comparative Analysis of SMP Click Scheduling Techniques. In: Ajmone Marsan M., Bianchi G., Listanti M., Meo M. (eds) Quality of Service in Multiservice IP Networks. QoS-IP 2004. Lecture Notes in Computer Science, vol 3375. Springer, Berlin, Heidelberg

Abstract

The interest of the scientific and commercial telecommunications community for the use of software routers running in general purpose (PC) hardware, as an alternative to the traditional special purpose hardware routers, is risen quickly in the last few years. This is due to the high level of flexibility and extensibility of this solution: the support for new protocols and network architectures and services, in fact, is easily obtained by re-programming the router itself. In addition, the diffusion of multiprocessor systems due to the progress in the semiconductor technologies allows software routers to obtain high performance if supported by multiprocessor PC hardware. Of course, in order to achieve a good use of the potentiality offered by multiprocessor architectures, the distribution of the tasks among the CPUs, and the parallel execution of the different operations, requires to be performed with some care. This paper demonstrates the benefits given by the hardware technological improvements, mainly concerning the use of multiple CPUs systems with respect to single processor ones, and shows how excellent forwarding performance can be achieved by Click software routers running on high powered PC hardware. Moreover, through the comparative analysis of different CPU scheduling approaches available in SMP Click, the paper discusses how different CPU scheduling techniques, that is, different approaches in the assignment of the tasks to the different CPUs, affect the router performance.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag Berlin Heidelberg 2005

Authors and Affiliations

  • Giorgio Calarco
    • 1
  • Carla Raffaelli
    • 1
  • Giovanni Schembra
    • 2
  • Giovanni Tusa
    • 2
  1. 1.D.E.I.S.University of BolognaBolognaItaly
  2. 2.D.I.I.T.University of CataniaCataniaItaly

Personalised recommendations