Abstract
In this paper we analyze a graph model representing the coarse grain dependency and communication structure of a distributed application. The model is called Timed Structural Parallelism Graph (TSPG). Nodes represent program components, arcs represent dependencies among components. This workload model differs from well known task graphs in two ways: (1) arcs can either have dependence or activation semantics and (2) timing parameters associated to arcs and nodes are given as intervals. Besides describing this new workload model, we sketch the issues and problems in corresponding evaluation techniques. In particular, we investigate techniques for estimating the total execution time and for deriving potential parallelism profiles. The proposed techniques are illustrated by example.
This work is supported by the Austrian National Science Foundation (FWF).
This is a preview of subscription content, log in via an institution.
Preview
Unable to display preview. Download preview PDF.
References
Markus Braun, Guenter Haring, and Gabriele Kotsis. Deriving parallelism profiles from structural parallelism graphs. In Proc. of the TDP'96, 1996.
Maria Calzarossa, Guenter Haring, Gabriele Kotsis, Alessandro Merlo, and Daniele Tessera. A hierarchical approach to workload characterization for parallel systems. In B. Hertzberger and G. Serazzi, Ed., High Performance Computing and Networking, LNCS vol. 919, pages 102–109, 1995.
Maria Calzarossa and Guiseppe Serazzi. Workload characterization: A survey. Proc. of the IEEE, 81(8):1136–1150, August 1993.
Franz Hartleb and Vassilis Mertsiotakis. Bounds for the mean runtime of parallel programs. In Rob Pooley and Jane Hillston, Eds., Proc. of the 6th Int. Conf. on Modelling Techniques and Tools for Computer Performance Evaluation, pages 197–210, 1992.
Johannes Lüthi, Gabriele Kotsis, Shikharesh Majumdar, and Günter Haring. Bounds-based performance analysis for distributed systems with variabilities and uncertainties in workload. In Proc. of DAPSYS'96, pages 51–58, Hungarian Academy of Sciences Report KFKI-1996-09/M,N, October 1996.
Shikharesh Majumdar and Revathy Ramadoss. Interval-based performance analysis of computing systems. In Patrick Dowd and Erol Gelenbe, Eds. Proc. MASCOTS 95, pages 345–351. IEEE CS Press, Jan. 1995.
Robin A. Sahner and Kishor S. Trivedi. SHARPE: Symbolic Hierarchical Automated Reliability and Performance Evaluator — Introduction and Guide for Users. Gould CSD, Urbana, 1101 E. University, Urbana, IL 61801, Sep. 1986.
W. M. P. van der Aalst. Using interval timed coloured petri nets to calculate performance bounds. available upon request from wsinwa@win.tue.nl, 1994.
C. Murray Woodside, Shikharesh Majumdar, and J. E. Neilson. Interval arithmetic for computing performance guarantees in client-server software. In F. Dehne, F. Fiala, and W. W. Koczkodaj, Eds. LNCS 497: Proc. ICCI '91, pages 535–546, Berlin, et al., 1991. Springer-Verlag.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1997 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Braun, M., Kotsis, G. (1997). Interval based workload characterization for distributed systems. In: Marie, R., Plateau, B., Calzarossa, M., Rubino, G. (eds) Computer Performance Evaluation Modelling Techniques and Tools. TOOLS 1997. Lecture Notes in Computer Science, vol 1245. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0022206
Download citation
DOI: https://doi.org/10.1007/BFb0022206
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-63101-9
Online ISBN: 978-3-540-69131-0
eBook Packages: Springer Book Archive