Abstract
Self-Adapting Numerical Software (SANS) systems aim to automate some of the laborious human decision making involved in adapting numerical algorithms to problem data, network conditions, and computational platform. In this paper we describe the structure of a SANS system that tackles automatic algorithm choice, based on dynamic inspection of the problem data. We describe the various components of such a system, and their interfaces.
| Original language | English |
|---|---|
| Pages (from-to) | 137-143 |
| Number of pages | 7 |
| Journal | International Journal of Parallel Programming |
| Volume | 33 |
| Issue number | 2-3 |
| DOIs | |
| State | Published - Jun 2005 |
| Externally published | Yes |
Funding
This research was supported in part by NSF NGS program under grant ACI-0203984 and in part by the AMS/MICS office of the DOE under contract DE-ACO5-00OR22725 with UT-Battelle, LLC., and in part by the Los Alamos Computer Science Institute through the subcontract # R71700J-29200099 from Rice University.
| Funders | Funder number |
|---|---|
| Los Alamos Computer Science Institute | R71700J-29200099 |
| National Science Foundation | ACI-0203984 |
| U.S. Department of Energy | DE-ACO5-00OR22725 |
| Rice University |
Keywords
- Adaptive systems
- Component frameworks
- Linear system solving