Distributed Application Monitoring for Clustered SMP Architectures

  • Karl Fürlinger
  • Michael Gerndt
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2790)


Performance analysis for terascale computing requires a combination of new concepts including distribution, on-line processing and automation. As a foundation for tools realizing these concepts, we present a distributed monitoring approach for clustered SMP architectures that tries to minimize the perturbation of the target application while retaining flexibility with respect to filtering and processing of performance data. We achieve this goal by dividing the monitor in a passive monitoring library linked to the application and an active component called runtime information producer (RIP) that provides performance data (metric- and event based) for individual nodes. Instead of adding an additional layer in the monitoring system that integrates performance data form the individual RIPs we include a directory service as a third component in our approach. Querying this directory service, tools discover which RIPs provide the data they need.


Performance Data Directory Service Address Space Target Application Monitoring Approach 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Balis, B., Bubak, M., Funika, W., Szepieniec, T., Wismüller, R.: Monitoring of Interactive Grid Applications. In: Proceedings of Dagstuhl Seminar 02341 on Performance Analysis and Distributed Computing. Kluwer Academic Publishers, Dordrecht (2003) (to appear)Google Scholar
  2. 2.
    Mohr, B., Malony, A.D., Shende, S., Wolf, F.: Towards a Performance Tool Interface for OpenMP: An Approach Based on Directive Rewriting. In: EWOMP 2001 Third European Workshop on OpenMP (September 2001)Google Scholar
  3. 3.
    CrossGrid Project ,
  4. 4.
    Dynamic Probe Class Library,
  5. 5.
    Dyninst. An Application Program Interface (API) for Runtime Code Generation,
  6. 6.
    Gerndt, M., Fürlinger, K.: Towards Automatic Performance Analysis for arge Scale Systems. In: At the 10th InternationalWorkshop on Compilers for Parallel Computers (CPC 2003), Amsterdam, The Netherlands (January 2003)Google Scholar
  7. 7.
    The Hitachi Performance Monitor Function (Hitachi System Documentation)Google Scholar
  8. 8.
    The Top 500 Supercomputer Sites,
  9. 9.
    Fahringer, T., Gerndt, M., Riley, G., Träff, J.L.: Formalizing OpenMP Performance Properties with the APART Specification Language (ASL). In: International Workshop on OpenMP: Experiences and Implementation, Tokyo, Japan, LNCS , pp. 428–439. Springer, Heidelberg (2000)Google Scholar
  10. 10.
    Fahringer, T., Gerndt, M., Riley, G., Träff, J.L.: Knowledge Specification for Automatic Performance Analysis. APART Technical Report (2001),
  11. 11.
    Ludwig, T., Wismüller, R., Sunderam, V., Bode, A.: OMIS – On-line Monitoring Interface Specification (Version 2.0). Shaker Verlag, Aachen Vol 9, LRR-TUM Research Report Series (1997),
  12. 12.
    Nagel, W.E., Arnold, A., Weber, M., Hoppe, H.C., Solchenbach, K.: VAMPIR: Visualization and analysis of MPI resources. Supercomputer 12(1), 69–80 (1996),

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • Karl Fürlinger
    • 1
  • Michael Gerndt
    • 1
  1. 1.Institut für Informatik, Lehrstuhl für Rechnertechnik und RechnerorganisationTechnische Universität München 

Personalised recommendations