Advertisement

Knowledge-Based Runtime Prediction of Stateful Web Services for Optimal Workflow Construction

  • Zoltan Balogh
  • Emil Gatial
  • Michal Laclavik
  • Martin Maliska
  • Ladislav Hluchy
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3911)

Abstract

This article proposes an approach for predicting runtime of web services (WS) with state – also called stateful web services. Estimating WS runtime is particularly critical during construction of composite WS workflows. Each workflow job must be scheduled in a way that the overall workflow run time will satisfy the overall workflow constrains. Such workflows are commonly used in Grids for connecting individual Grid WS to large, complicated and distributed applications. Prediction of WS run times optimizes scheduling and supports efficient use of grid resources. In our approach we propose to estimate expected WS run time based on invocation parameters of WS operations, states of resources maintained by a WS and properties of resources used as processing inputs for a WS. We adopt knowledge based approach where the history of WS operations is examined and a model is created and updated for each class and instance of a WS. Such WS run time prediction models can be then used by workflow schedulers to compute expected run times of a range of WS for the purpose of identifying the most appropriate WS for a given job within given constrains.

Keywords

Grid Resource Past Case Grid Project Weighted Euclidean Distance Grid Resource Usage 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
  2. 2.
    Web Service Management: Service Life Cycle: http://www.w3.org/TR/2004/NOTE-wslc-20040211/
  3. 3.
    EU IST: K-Wf Grid Project IST-2002-511385, http://www.kwfgrid.net/
  4. 4.
    EU IST: CrossGrid Project IST-2001-32243, http://www.crossgrid.org/
  5. 5.
    Delpoi Grid(Lab)? Adaptive Component System: http://www.gridlab.org/WorkPackages/wp-7/
  6. 6.
    Faerman, M., Su, A., Wolski, R., Berman, F.: Adaptive Performance Prediction for Distributed Data-Intensive Applications. In: Proceedings of the ACM/IEEE SC 1999 Conference on High Performance Networking and Computing, Portland, August 9 (1999)Google Scholar
  7. 7.
    Balogh, Z., Laclavik, M., Hluchy, L., Nguyen, T.G., Gatial, E.: Capture, Discovery and Reuse of Knowledge in REMARK. In: ICETA 2004, Kosice, Slovakia, September 2004, IEEE Computer Society, Los Alamitos (2004)Google Scholar
  8. 8.
    Menasce, D.A., Virgilio, A.F.: Capacity Planning for Web Services - Metrics, Models, and Methods. Prentice-Hall, Englewood Cliffs (2002)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Zoltan Balogh
    • 1
  • Emil Gatial
    • 1
  • Michal Laclavik
    • 1
  • Martin Maliska
    • 1
  • Ladislav Hluchy
    • 1
  1. 1.Institute of InformaticsSlovak Academy of SciencesBratislavaSlovakia

Personalised recommendations