Performance Analysis of a Parallel Application in the GRID
Performance analysis of real applications in clusters and GRID like environments is essential to fully exploit the performance of new architectures. The key problem is the deepening hierarchy of latencies and bandwidths and the growing heterogeneity of systems. This paper discusses the basic problems of performance analysis in such clustered and heterogeneous environments. It further presents a software environment that supports the user in running codes and getting more insight into the performance of the application. In order to give a proof of the concept a code for direct numerical simulation of reactive flows is run in a GRID like hardware environment, and the performance analysis is presented.
Unable to display preview. Download preview PDF.
- 1.R. Buyya. High Performance Cluster Computing, Volume 1: Architectures and Systems, Volume 2: Programming and Applications. Prentice Hall, 1999.Google Scholar
- 2.Ian Foster and Carl Kesselman. The Grid: Blueprint for a New Computing Infrastructure. Morgan Kaufmann, 1999.Google Scholar
- 5.H. Brunst, M. Winkler, W. E. Nagel, and H.-C. Hoppe. Performance Optimization for Large Scale Computing: The Scalable VAMPIR Approach. In V. N. Alexandrov, J. J. Dongarra, B. A. Juliano, R. S. Renner, and C. K. Tan, Eds., Computational Science-ICCS 2001, Part II, LNCS 2074, 751–760, Springer, 2001.CrossRefGoogle Scholar
- 6.S. Moore, D. Cronk, K. London, and J. Dongarra. Review of Performance Analysis Tools for MPI Parallel Programs. In Y. Cotronis and J. Dongarra, Eds., Recent Advances in Parallel Virtual Machine and Message Passing Interface, 8th European PVM/MPI Users’ Group Meeting, Proceedings, LNCS 2131, 241–248, Springer, 2001.CrossRefGoogle Scholar
- 8.J. Almond, D. Snelling. UNICORE: Secure and Uniform Access to Ditributed Resources via the World Wide Web. 1998. http://www.fz-juelich.de/unicore/whitepaper.ps.
- 9.METODIS-Metacomputing Tools for Distributed Systems. http://www.hlrs.de/organization/pds/projects/metodis.
- 10.DAMIEN-Distributed Applications and Middleware for Industrial Use of European Networks. http://www.hlrs.de/organization/pds/projects/damien.
- 11.F. A. Williams. Combustion Theory. second edition, Benjamin/Cummings, 1985.Google Scholar
- 12.M. Lange and J. Warnatz. Investigation of Chemistry-Turbulence Interactions Using DNS on the Cray T3E. In E. Krause and W. Jäger, Eds., High Performance Computing in Science and Engineering’ 99, 333–343, Springer, 2000.Google Scholar
- 13.M. Lange. Parallel DNS of Autoignition Processes with Adaptive Computation of Chemical Source Terms. In C. B. Jenssen, T. Kvamsdal, H. I. Andersson, B. Pettersen, A. Ecer, J. Periaux, N. Satofuka, and P. Fox, Eds., Parallel Computational Fluid Dynamics: Trends and Applications, 551–558, Elsevier Science, 2001.Google Scholar
- 14.M. Lange. Massively Parallel DNS of Flame Kernel Evolution in Spark-Ignited Turbulent Mixtures. In E. Krause and W. Jäger, Eds., High Performance Computing in Science and Engineering’ 02, 425–438, Springer, 2002.Google Scholar
- 15.E. Gabriel, M. Lange and R. Rühle. Direct Numerical Simulation of Turbulent Reactive Flows in a Metacomputing Environment. In Proceedings of the 2001 ICPP Workshops, 237–244, IEEE Computer Society, 2001.Google Scholar
- 16.E. Gabriel, M. Resch and R. Rühle. Implementing MPI with Optimized Algorithms for Metacomputing. In A. Skjellum, P. V. Bangalore, Y. S. Dandass, Eds., Proceedings of the Third MPI Developer’s and User’s Conference, MPI Software Technology Press, 1999.Google Scholar
- 17.Message Passing Interface Forum. MPI: A Message-Passing Interface Standard (Version 1.1). 1995. http://www.mpi-forum.org.