Abstract
Given an algorithm and architecture a tuning parameter is an input parameter that has consequences in the performance but not in the output. The list of tuning parameters in parallel computing is extensive: some depending on the architecture, as the number of processors and the size of the buffers used during data exchange and some depending on the application. We formalize the General Tuning Problem and propose a generic methodology to solve it. The technique is applied to the special case of pipeline algorithms. A tool that automatically solves the prediction of the tuning parameters is presented. The accuracy is tested on a CRAY T3E. The results obtained suggest that the technique could be successfully ported to other paradigms.
The work described in this paper has been partially supported by the Spanish Ministry of Science and Technology (CICYT) TIC1999-0754-C03.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Andonov R., Rajopadhye S.. Optimal Orthogonal Tiling of 2D Iterations. Journal of Parallel and Distributed computing, 45(2), (1997) 159–165.
Culler D., Karp R., Patterson D., Sahay A., Schauser K., Santos E., Subramonian R., von Eicken T.. LogP: Towards a realistic model of parallel computaiotn. Proceeings of the 4th ACM SIGPLAN. Sym. Principles nad Practice of Parallel Programming. May 1993.
Espinosa A., Margalef T., Luque E.. Automatic Performance Evaluation of Parallel Programs. Proc. Of the 6th EUROMICRO Workshop on Parallel and Distributed Processing. IEEE CS. 1998. 43–49.
Fahringer T., Zima H.. Static Parameter Based Performance Prediction Tool for Parallel Programs. Proc. of ACM International Conference of Supercomputing. ACM Press. 1993. 207–219.
Fortune S., Wyllie J. Parallelism in Randomized Machines. Proceedings of STOC. 1978. 114–118.
Geist A., Heath M., Peyton B., Worley P.. PICL: Aportable Instrumented Communications Lybrary, C Reference Manual. Technical Report TM-11130. Oak Ridge National Laboratory. 1990.
Heath M. Etheridge J.. Visualizing the Performance of Parallel Programs. IEEE Software. 8(5). September 1991. 29–39.
Labarta J., Girona S., Pillet V., Cortes T., Gregoris L.. Dip: A Parallel Program Development Environment. Europar 96. Lyon. August 1996.
Morales D., Almeida F., García F., González J., Roda J., Rodríguez C.. A Skeleton for Parallel Dynamic Programming. Euro-Par’99 Parallel Processing Lecture Notes in Computer Science, Vol. 1685. Springer-Verlag, (1999) 877–887.
Morales D., Almeida F., Moreno L. M., Rodríguez C.. Optimal Mapping of Pipeline Algorithms. EuroPar 2000. Munich. Sept. 2000. 320–324
Roda J., Rodriguez C., Gonzalez D., Almeida F.. Title: Predicting the Execution Time of Message Passing Models. Concurrency: Practice & Experience. Addison Wesley. Vol 11(9). 1999. 461–447
Rodriguez C., Roda J., Garcia F., Almeida F., Gonzalez D.. Paradigms for Parallel Dynamic Programming. Proceedings of the 22nd Euromicro Conference. Beyond 2000: Hardware and Software Strategies. IEEE. (1996) 553–563.
Valiant L. A Bridging Model for Parallel Computation. Commun. ACM, 33(8). 1990. 103–111.
Vampir 2.0 Visualization and Analysis of MPI Programs. http://www.pallas.com
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2001 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Moreno, L.M., Almeida, F., González, D., Rodríguez, C. (2001). Adaptive Execution of Pipelines. In: Cotronis, Y., Dongarra, J. (eds) Recent Advances in Parallel Virtual Machine and Message Passing Interface. EuroPVM/MPI 2001. Lecture Notes in Computer Science, vol 2131. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-45417-9_31
Download citation
DOI: https://doi.org/10.1007/3-540-45417-9_31
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-42609-7
Online ISBN: 978-3-540-45417-5
eBook Packages: Springer Book Archive