Simulation of Job Execution in the M5PS Multiprocessor

  • F. Regen
  • L. Krings
  • W. Ameling
Conference paper
Part of the Informatik-Fachberichte book series (INFORMATIK, volume 71)


Resource sharing in multiprocessor systems has a strong influence on the speed-up of the processing elements. Analytic computational methods, e.g. queueing theory, do not provide time-dependent solutions by tolerable effort,’ whereas simulation often requires a great deal of computational time and is more difficult to validate. In order to overcome these disadvantages, the paper presents a model that combines analytic computation and simulation. The time-dependent solution of the model under time-dependent load is obtained by the simulation package FORCASD. For reasons of state reduction and decrease of computational effort the hardware layer is modeled by a queueing model. The results are validated by measurements in the M5PS multiprocessor system.


Shared Memory Multiprocessor System User Program Parent Process Tolerable Effort 
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. /BASK71/.
    Baskett, F.: The Dependence of Computer System Queues upon Processing Time Distribution and Central Processor Scheduling Proceedings ACM SIGOPS, 3rd Symp. on Operating System Principles, Stanford University, 10/1971Google Scholar
  2. /DAHM82/.
    Dahmen, N.:Modelling and Simulation with FORCASD - User Manual Technical Report LB 528/82, Philips GmbH, Forschungslaboratorium Hamburg, 1982Google Scholar
  3. /DAHM83/.
    Dahmen, N.: FORCASD - An Evaluation Net Oriented Program System for Modelling and Simulation Proceedings ESC83, Aachen, 9/1983Google Scholar
  4. /HOEN77/.
    Hoener, S.;Roehder, W.: Efficiency of a Multi-Microprocessor System with Time-Shared Buses Proceedings EUROMICRO 77, 1977Google Scholar
  5. /KLE176/.
    Kleinrock, L.: Queueing Systems Vol. 1 Wiley & Sons, New York, 1976MATHGoogle Scholar
  6. /KRIN81/.
    Krings, L.; Milde, J.; Ameling, W.: An Approach to Performance Measuring in Multiprocessor Systems with Time-Shared Buses Proceedings EUROMICRO 81, Paris, 9/1981Google Scholar
  7. /KRIN82/.
    Krings, L.;Milde, J.;Ameling, W.: The Influence of Bus Allocation Algorithms on The System Performance of Multiprocessor Systems with a Time-Shared Bus Proceedings 10th IMACS World Congress, Montreal, 8/1982Google Scholar
  8. /MILD82a/.
    Milde, J.;Krings, L.;Ameling, W.: Architektur des Multiprozessor Systems M5PS und Auswertung einiger Anwendungen NTG/GI-Fachtagung, Ulm, 3/1982Google Scholar
  9. /MILD82b/.
    Milde, J.;Krings, L.;Ameling, W.: Realization of Synchronization Tools and Their Efficiency in the Multiprocessorsystem M5PS Proceedings 10th IMACS World Congress, Montreal, 8/1982Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1983

Authors and Affiliations

  • F. Regen
    • 1
  • L. Krings
    • 1
  • W. Ameling
    • 1
  1. 1.Aachen Technical UniversityGermany

Personalised recommendations