An Approach to Performance Prediction for Parallel Applications
Accurately modeling and predicting performance for large-scale applications becomes increasingly difficult as system complexity scales dramatically. Analytic predictive models are useful, but are difficult to construct, usually limited in scope, and often fail to capture subtle interactions between architecture and software. In contrast, we employ multilayer neural networks trained on input data from executions on the target platform. This approach is useful for predicting many aspects of performance, and it captures full system complexity. Our models are developed automatically from the training input set, avoiding the difficult and potentially error-prone process required to develop analytic models. This study focuses on the high-performance, parallel application SMG2000, a much studied code whose variations in execution times are still not well understood. Our model predicts performance on two large-scale parallel platforms within 5%-7% error across a large, multi-dimensional parameter space.
- Research Organization:
- Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States)
- Sponsoring Organization:
- USDOE
- DOE Contract Number:
- W-7405-ENG-48
- OSTI ID:
- 878233
- Report Number(s):
- UCRL-CONF-212365; TRN: US200611%%86
- Resource Relation:
- Journal Volume: 3648; Conference: Presented at: Euro-Par 2005, Lisboa, Portugal, Portugal, Aug 30 - Sep 02, 2005
- Country of Publication:
- United States
- Language:
- English
Semicoarsening Multigrid on Distributed Memory Machines
|
journal | January 2000 |
hypre: A Library of High Performance Preconditioners
|
book | April 2002 |
A First-Order Superscalar Processor Model
|
journal | March 2004 |
Predictive performance and scalability modeling of a large-scale application
|
conference | January 2001 |
A Performance Prediction Framework for Scientific Applications
|
book | January 2003 |
Cross-architecture performance predictions for scientific applications using parameterized models
|
conference | January 2004 |
Similar Records
Computer Science Research Needs for Parallel Discrete Event Simulation (PDES)
Fouling resistance prediction using artificial neural network nonlinear auto-regressive with exogenous input model based on operating conditions and fluid properties correlations