A Hybrid Approach for Performance Evaluation of Web-Legacy Client/Server Systems

  • Naoki Makimoto
  • Hiroyuki Sakata
Conference paper

Abstract

In this paper, we propose a hybrid approach to evaluate the performance of Web-legacy client/server systems. The approach combines three methods, hardware and software simulations and approximate analysis. Hardware simulation measures basic performance characteristics of the system whose data are utilised in the subsequent methods for designing the system so as to satisfy the prescribed quality of service. From our experience of its application to a prototype system, we observed that both software simulation and approximate analysis provide reasonable evaluation of the system performance.

Key words

simulation Web-legacy C/S system performance evaluation quality of service 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Chao, X., Miyazawa, M. and Pinedo, M. (1999) Queueing Networks. John Wiley & Sons.Google Scholar
  2. 2.
    Gelenbe, E. and Mitrani, I. (1980) Analysis and Synthesis of Computer Systems. Academic Press.Google Scholar
  3. 3.
    Kimura, T. (1994) Approximations for multi-server queues: system interpolations. Queueing Systems, 17(3–4), 347–382.MATHCrossRefGoogle Scholar
  4. 4.
    Kino, I. (2002) Queueing Networks. Asakura Syoten (in Japanese).Google Scholar
  5. 5.
    Menascé, D.A. and Almeida, V.A.F. (1998) Capacity Planning for Web Performance. Prentice Hall PTR.Google Scholar
  6. 6.
    Menascé, D.A. and Almeida, V.A.F. (2002) Capacity Planning for Web Services. Prentice Hall PTR.Google Scholar
  7. 7.
    Pritsker, A.A.B., O’Reilly, J.J. and LaVal, D.K. (1997) Simulation with Visual SLAM and Awesim. Systems Publishing and John Wiley.Google Scholar
  8. 8.
    Ripley, B.D. (1987) Stochastic Simulation. Wiely.Google Scholar
  9. 9.
    Standard Performance Evaluation Corporation (2002) SPECjAppServer2002 Documentation. http://www.spec.org/osg/jAppServer.Google Scholar
  10. 10.
    Sun Microsystems, Inc. Enterprise Java Beans 1.1 and 2.0. Specifications. http://java.sun.com/products/ejbGoogle Scholar

Copyright information

© Springer 2008

Authors and Affiliations

  • Naoki Makimoto
    • 1
  • Hiroyuki Sakata
    • 2
  1. 1.University of TsukubaJapan
  2. 2.NTT Data CorporationJapan

Personalised recommendations