Published November 18, 2020
| Version v1
Conference paper
Open
Applying StarPU runtime system to scientific applications: Experiences and lessons learned
Creators
- 1. NTUA - National Technical University of Athens [Athens]
- 2. MDLS - Maison de la Simulation
- 3. LIUPPA - Laboratoire Informatique de l'Université de Pau et des Pays de l'Adour
- 4. Linköpings universitet
- 5. STORM - STatic Optimizations, Runtime Methods LaBRI - Laboratoire Bordelais de Recherche en Informatique, Inria Bordeaux - Sud-Ouest
- 6. Forschungszentrum Jülich GmbH
Description
Task-based runtime systems are adopted by application developers for their valuable features including flexibility of execution and optimized resource management. However, the use of such advanced programming models in complex HPC applications often requires significant training time and programming effort. In this work, we share experiences and lessons learned from the use of StarPU in three independent projects of various complexity. We reach conclusions, with respect to training, programming effort, and existing challenges, that are useful to the communities of application developers, as well as to the developers of runtime systems. Finally, we suggest extensions to the runtime systems beneficial to application developers.
Files
publication-7.pdf
Files
(807.5 kB)
| Name | Size | Download all |
|---|---|---|
|
md5:b2b79dbeb3a261cc2bfabab17452e226
|
807.5 kB | Preview Download |