A Framework for Performance Modeling and Prediction
- 1 January 2002
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 2 (10639535) , 21
- https://doi.org/10.1109/sc.2002.10004
Abstract
Cycle-accurate simulation is far too slow for modeling the expected performance of full parallel applications on large HPC systems. And just running an application on a system and observing wallclock time tells you nothing about why the application performs as it does (and is anyway impossible on yet-to-be-built systems). Here we present a framework for performance modeling and prediction that is faster than cycle-accurate simulation, more informative than simple benchmarking, and is shown useful for performance investigations in several dimensions.Keywords
This publication has 3 references indexed in Scilit:
- Modeling application performance by convolving machine signatures with application profilesPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2005
- Accurate performance prediction for massively parallel systems and its applicationsPublished by Springer Nature ,1996
- OCCAM'S RAZORMind, 1915