StarPU-MPI: Task programming over clusters of machines enhanced with accelerators

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

Volumn 7490 LNCS, Issue , 2012, Pages 298-299

  Augonnet, Cédric ^a   Aumage, Olivier ^c   Furmento, Nathalie ^b   Namyst, Raymond ^b   Thibault, Samuel ^b  

^a NVIDIA   (United States)

^b UNIVERSITÉ DE BORDEAUX   (France)

^c INRIA   (France)

Author keywords

Accelerators; GPUs; MPI; Task based model

Indexed keywords

GPUS; HETEROGENEOUS NODES; MPI; RUNTIME SYSTEMS; SCHEDULING TASKS; TASK GRAPH; TASK PROGRAMMING; TASK-BASED;

ARTIFICIAL INTELLIGENCE; PARTICLE ACCELERATORS;

PROGRAM PROCESSORS;

EID: 84867630228     PISSN: 03029743     EISSN: 16113349     Source Type: Book Series    
DOI: 10.1007/978-3-642-33518-1_40     Document Type: Conference Paper

References (2)

1. Augonnet, C., Thibault, S., Namyst, R., Wacrenier, P.A.: StarPU: A Unified Platform for Task Scheduling on Heterogeneous Multicore Architectures. In: Sips, H., Epema, D., Lin, H.-X. (eds.) Euro-Par 2009. LNCS, vol. 5704, pp. 863-874. Springer, Heidelberg (2009)
2. Augonnet, C., Clet-Ortega, J., Thibault, S., Namyst, R.: Data-Aware Task Scheduling on Multi-Accelerator based Platforms. In: The 16th International Conference on Parallel and Distributed Systems, ICPADS (2010)