The documents distributed by this server have been provided by the contributing authors as a means to ensure timely dissemination of scholarly and technical work on a noncommercial basis. Copyright and all rights therein are maintained by the authors or by other copyright holders, notwithstanding that they have offered their works here electronically. It is understood that all persons copying this information will adhere to the terms and constraints invoked by each author's copyright. These works may not be reposted without the explicit permission of the copyright holder.
Publications of SPCL
|W. Gropp, T. Hoefler, M. Snir:|
|Performance Modeling for Systematic Performance Tuning|
(Presentation - In SIAM Conference on Computational Science and Engineering 2011 (Abstracts), presented in Reno, NV, SIAM, ISSN: , ISBN: , Feb. 2011, )
AbstractThe performance of parallel scientific applications depends on many factors which are determined by the execution environment and the parallel application. Especially on large parallel systems, it is too expensive to explore the solution space with series of benchmarks. Deriving analytical models for applications and platforms allow estimating and extrapolating their execution performance, bottlenecks, and the potential impact of optimization options. We propose to use such "performance modeling" techniques beginning from the design process throughout the whole software development cycle. We argue that performance models be maintained and updated during the life of a code. Such models help to guide design decisions and re-engineering efforts to adopt applications to changing platforms (e.g., GPU or multicore computing) and allow users to estimate costs to solve a particular problem. Application performance models can be defined at different levels of abstraction beginning from simple asymptotic models that allow rough statements about the scaling behavior with respect to specific input arguments to fully paraterized models that allow absolute time predictions on a particular architecture. Models can often be built with the help of well-known performance profiling tools. We will motivate the use of performance modeling with examples from the Blue Waters project.