 |
|
 |
|
[Back to Research topics] |
Performance Prediction using Machine Learning to speed-up Design Space Exploration (DSE)
|
Description
Collaborations
Software
Our related publications
Links
|
|
Description: |
I am moving all my research and software development activities to the UNIDAPT Group.
This page will give you some information about my previous activities.
Also, my personal research page can give you some more information about my current activities.
 
Developing an optimizing compiler for new architectures is extremely
difficult and time-consuming when there is only a simulator of the machine available. We suggest to automatically build a performance model for a new architecture to predict the impact of different program transformations on this architecture using statistical and machine learning techniques, and a limited number of training runs. We use either statical or run-time features of the program to quickly predict the performance of all transformations for a given program on a new architecture thus speeding up the evaluation of compiler/architecture performance by several orders of magnitude. We also developed a novel technique to characterize programs or architectures using program reaction to optimizations (transformations).
|
| Software: |
Collaborations: |
More Information about my software developments
|
|
|
Our related publications: |
-
Grigori Fursin, Cupertino Miranda, Olivier Temam, Mircea Namolaru, Elad Yom-Tov, Ayal Zaks, Bilha Mendelson, Phil Barnard, Elton Ashton, Eric Courtois, Francois Bodin, Edwin Bonilla, John Thomson, Hugh Leather, Chris Williams, Michael O'Boyle. MILEPOST GCC: machine learning based research compiler. Proceedings of the GCC Developers' Summit, Ottawa, Canada, June 2008
[bib] [pdf]
-
Veerle Desmet, Grigori Fursin, Sylvain Girbal and Olivier Temam. Leveraging Modular Simulation for Systematic Design Space Exploration. 4th HiPEAC Industrial Workshop on Compilers and Architectures organized by ARM Ltd., Cambridge, UK, November 2007
[bib]
- Grigori Fursin and Albert Cohen. Building a Practical Iterative Interactive Compiler. 1st International Workshop on Statistical and Machine Learning Approaches Applied to Architectures and Compilation (SMART'07), Ghent, Belgium, January 2007
[bib] [pdf]
- John Cavazos, Christophe Dubach, Felix Agakov, Edwin Bonilla, Michael F.P. O'Boyle, Grigori Fursin and Olivier Temam. Automatic Performance Model Construction for the Fast Software Exploration of New Hardware Designs. International Conference on Compilers, Architecture, And Synthesis For Embedded Systems (CASES 2006), Seoul, Korea, October 2006
(finalist best paper award)
[bib] [pdf]
All my publications My bibliography collection
My links to on-line journals and conference
|
|
Links: |
- will add more groups soon ...
If I forgot some group, please don't hesitate to contact me.
|
|
|
|
|
|
|
|
