Abstract
In recent years, automatic data-driven modeling with machine learning (ML) has received considerable attention as an alternative to analytical modeling for many modeling tasks. While ad hoc adoption of ML approaches has obtained success, the real potential for automation in data-driven modeling has yet to be achieved. We propose AutoMOMML, an end-to-end, ML-based framework to build predictive models for objectives such as performance, and power. The framework adopts statistical approaches to reduce the modeling complexity and automatically identifies and configures the most suitable learning algorithm to model the required objectives based on hardware and application signatures. The experimental results using hardware counters as application signatures show that the median prediction error of performance, processor power, and DRAM power models are 13 %, 2.3%, and 8%, respectively.
Original language | English |
---|---|
Title of host publication | High Performance Computing - 31st International Conference, ISC High Performance 2016, Proceedings |
Editors | Jack Dongarra, Julian M. Kunkel, Pavan Balaji |
Publisher | Springer Verlag |
Pages | 219-239 |
Number of pages | 21 |
ISBN (Print) | 9783319413204 |
DOIs | |
State | Published - 2016 |
Externally published | Yes |
Event | 31st International Conference on High Performance Computing, ISC High Performance 2016 - Frankfurt, Germany Duration: Jun 19 2016 → Jun 23 2016 |
Publication series
Name | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
---|---|
Volume | 9697 |
ISSN (Print) | 0302-9743 |
ISSN (Electronic) | 1611-3349 |
Conference
Conference | 31st International Conference on High Performance Computing, ISC High Performance 2016 |
---|---|
Country/Territory | Germany |
City | Frankfurt |
Period | 06/19/16 → 06/23/16 |
Funding
This work was supported by the U.S. Department of Energy, Office of Science, Advanced Scientific Computing Research program under contract number DE-AC02-06CH11357.