Predictive performance and scalability modeling of a large-scale application

作者: D. J. Kerbyson , H. J. Alme , A. Hoisie , F. Petrini , H. J. Wasserman

DOI: 10.1145/582034.582071

关键词:

摘要: In this work we present a predictive analytical model that encompasses the performance and scaling characteristics of an important ASCI application. SAGE (SAIC's Adaptive Grid Eulerian hydrocode) is multidimensional hydrodynamics code with adaptive mesh refinement. The validated against measurements on several systems including Blue Mountain, White, Compaq Alphaserver ES45 system showing high accuracy. It parametric --- basic machine numbers (latency, MFLOPS rate, bandwidth) application (problem size, decomposition method, etc.) serve as input. applied to add insight into current systems, reveal bottlenecks, illustrate where tuning efforts can be effective. We also use predict future systems.

参考文章(10)
Report, Michael B secretary, Public International Benchmarks for Parallel Computers University of Tennessee. ,(1993)
Adolfy Hoisie, Stefan Goedecker, Performance Optimization of Numerically Intensive Codes ,(1987)
Silvia Müller, Per Stenström, Mateo Valero, Stamatis Vassiliadis, Parallel Computer Architecture european conference on parallel processing. pp. 537- 538 ,(2000) , 10.1007/3-540-44520-X_72
G. R. Nudd, D. J. Kerbyson, E. Papaefstathiou, S. C. Perry, J. S. Harper, D. V. Wilcox, Pace--A Toolset for the Performance Prediction of Parallel and Distributed Systems ieee international conference on high performance computing data and analytics. ,vol. 14, pp. 228- 251 ,(2000) , 10.1177/109434200001400306
Patrick H. Worley, Performance evaluation of the IBM SP and the Compaq AlphaServer SC international conference on supercomputing. pp. 235- 244 ,(2000) , 10.1145/335231.335254
Adolfy Hoisie, Olaf Lubeck, Harvey Wasserman, Performance and Scalability Analysis of Teraflop-Scale Parallel Architectures Using Multidimensional Wavefront Applications ieee international conference on high performance computing data and analytics. ,vol. 14, pp. 330- 346 ,(2000) , 10.1177/109434200001400405
F. Petrini, Wu-chun Feng, A. Hoisie, S. Coll, E. Frachtenberg, The Quadrics network (QsNet): high-performance clustering technology high performance interconnects. pp. 125- 130 ,(2001) , 10.1109/HIS.2001.946704
T. Rauber, G. Runger, Modelling the runtime of scientific programs on parallel computers international workshop on parallel processing. pp. 307- 314 ,(2000) , 10.1109/ICPPW.2000.869117
John B. Drake, Steve Hammond, Rodney James, Patrick H. Worley, Performance Tuning and Evaluation of a Parallel Community Climate Model conference on high performance computing (supercomputing). pp. 34- 34 ,(1999) , 10.1145/331532.331566