Abstract
Scientific and mathematical parallel libraries offer a high level of abstraction to programmers. However, it is still difficult to select the proper parameters and algorithms to maximize the application performance. This work proposes a performance model for dynamically adjusting applications written with the PETSc library. This model is based on historical performance information and data mining techniques. Finally, we demonstrate the validity of the proposed model through real experimentations.
Original language | American English |
---|---|
Pages (from-to) | 110-119 |
Number of pages | 10 |
Journal | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
DOIs | |
Publication status | Published - 2008 |
Keywords
- Dynamic mathematical model
- Mathematical Performance
- Performance Model
- PETSc Performance