 |
|
|
|
| HAL : hal-00359342, version 4 |
| Fiche détaillée |  Récupérer au format |
|
|
| Versions disponibles : | v1 (06-02-2009) | v2 (09-04-2009) | v3 (21-09-2009) | v4 (24-09-2009) |
|
|
|
|
| Barra, a Parallel Functional GPGPU Simulator |
|
|
| Sylvain Collange 1David Defour 1 |
|
|
| (06/02/2009) |
|
|
|
| We present a GPU functional simulator targeting GPGPU based on the UNISIM framework which takes unaltered NVIDIA CUDA executables as input. It simulates the native instruction set of the Tesla architecture at the functional level and generates detailed execution statistics. Simulation speed is competitive with the less-accurate CUDA emulation mode thanks to optimizations which exploit the inherent parallelism of GPGPU applications. Furthermore, it opens the way for GPU microarchitecture design space exploration. |
|
|
|
|
|
|
|
|
|
|
|
| 1 : | Laboratoire de Physique Appliquée et d'Automatique (LP2A) |
|
|
|
EA3679 – Université de Perpignan |
|
|
|
|
|
|
|
|
|
| Domaine | : | Informatique/Architecture Informatique/Calcul formel Informatique/Performance et fiabilité Informatique/Modélisation et simulation |
|
|
| GPGPU – CUDA – UNISIM – simulation – SIMD |
|
|
|
| Liste des fichiers attachés à ce document : | |||||
|
|||||
|
|
|
| hal-00359342, version 4 | |
| http://hal.archives-ouvertes.fr/hal-00359342 | |
| oai:hal.archives-ouvertes.fr:hal-00359342 | |
| Contributeur : Sylvain Collange | |
| Soumis le : Jeudi 24 Septembre 2009, 16:36:13 | |
| Dernière modification le : Jeudi 24 Septembre 2009, 16:49:51 | |
