Publications of year 2017
Table of Contents
Thesis
Arthur Chevalier
Critical resources management and scheduling under StarPU
Master Thesis, Université de Bordeaux, September 2017
[WWW] [PDF] Keyword(s): On Memory Control, On Hierarchical Tasks
@phdthesis{chevalier:hal-01718280, TITLE = {{Critical resources management and scheduling under StarPU}}, AUTHOR = {Chevalier, Arthur}, URL = {https://hal.inria.fr/hal-01718280}, SCHOOL = {{Universit{\'e} de Bordeaux}}, YEAR = {2017}, MONTH = Sep, KEYWORDS = {On Memory Control; On Hierarchical Tasks}, PDF = {https://hal.inria.fr/hal-01718280/file/Memoire.pdf}, TYPE = {Master Thesis}, HAL_ID = {hal-01718280}, HAL_VERSION = {v1}, }
Suraj Kumar
Scheduling of Dense Linear Algebra Kernels on Heterogeneous Resources
PhD thesis, Université de Bordeaux, April 2017
[WWW] [PDF] Keyword(s): On Scheduling, STARPU, Runtime Systems, Heterogeneous Platforms, Task-based Scheduling, Dynamic Schedulers, Dense Linear Algebra, Systèmes d'ordonnancement dynamiques, Plates-formes hétérogènes, Algèbre linéaire dense, Ordonnancement dynamique, Ordonnancement à base de graphe de tâches
@phdthesis{kumar:tel-01538516, TITLE = {{Scheduling of Dense Linear Algebra Kernels on Heterogeneous Resources}}, AUTHOR = {Kumar, Suraj}, URL = {https://tel.archives-ouvertes.fr/tel-01538516}, NUMBER = {2017BORD0572}, SCHOOL = {{Universit{\'e} de Bordeaux}}, YEAR = {2017}, MONTH = Apr, KEYWORDS = {On Scheduling; STARPU ; Runtime Systems ; Heterogeneous Platforms ; Task-based Scheduling ; Dynamic Schedulers ; Dense Linear Algebra ; Syst{\`e}mes d'ordonnancement dynamiques ; Plates-formes h{\'e}t{\'e}rog{\`e}nes ; Alg{\`e}bre lin{\'e}aire dense ; Ordonnancement dynamique ; Ordonnancement {\`a} base de graphe de t{\^a}ches}, PDF = {https://tel.archives-ouvertes.fr/tel-01538516/file/KUMAR_SURAL_2017.pdf}, HAL_ID = {tel-01538516}, HAL_VERSION = {v1}, }
Articles in journal, book chapters
Emmanuel Agullo, Olivier Aumage, Berenger Bramas, Olivier Coulaud, and Samuel Pitoiset
Bridging the gap between OpenMP and task-based runtime systems for the fast multipole method
IEEE Transactions on Parallel and Distributed Systems, April 2017
[WWW] [PDF] [doi:10.1109/TPDS.2017.2697857] Keyword(s): On OpenMP Support on top of StarPU, STARPU, multicore architecture, commutativity, priority, high performance computing, fast multipole method, runtime system, OpenMP, compiler, parallel programming model, StarPU, KStar, ScalFMM
@article{agullo:hal-01517153, TITLE = {{Bridging the gap between OpenMP and task-based runtime systems for the fast multipole method}}, AUTHOR = {Agullo, Emmanuel and Aumage, Olivier and Bramas, Berenger and Coulaud, Olivier and Pitoiset, Samuel}, URL = {https://hal.inria.fr/hal-01517153}, JOURNAL = {{IEEE Transactions on Parallel and Distributed Systems}}, YEAR = {2017}, MONTH = Apr, DOI = {10.1109/TPDS.2017.2697857}, KEYWORDS = {On OpenMP Support on top of StarPU; STARPU ; multicore architecture ; commutativity ; priority ; high performance computing ; fast multipole method ; runtime system ; OpenMP ; compiler ; parallel programming model ; StarPU ; KStar ; ScalFMM}, PDF = {https://hal.inria.fr/hal-01517153/file/tpds_kstar_scalfmm_print.pdf}, HAL_ID = {hal-01517153}, HAL_VERSION = {v1}, }
Emmanuel Agullo, Olivier Aumage, Mathieu Faverge, Nathalie Furmento, Florent Pruvost, Marc Sergent, and Samuel Thibault
Achieving High Performance on Supercomputers with a Sequential Task-based Programming Model
TPDS - IEEE Transactions on Parallel and Distributed Systems, December 2017
[WWW] [PDF] [doi:10.1109/TPDS.2017.2766064] Keyword(s): On MPI Support, runtime system, sequential task flow, task-based programming, heterogeneous computing, distributed computing, multicore, GPU, Cholesky factorization
@article{agullo:hal-01618526, TITLE = {{Achieving High Performance on Supercomputers with a Sequential Task-based Programming Model}}, AUTHOR = {Agullo, Emmanuel and Aumage, Olivier and Faverge, Mathieu and Furmento, Nathalie and Pruvost, Florent and Sergent, Marc and Thibault, Samuel}, URL = {https://hal.inria.fr/hal-01618526}, JOURNAL = {{TPDS - IEEE Transactions on Parallel and Distributed Systems}}, PUBLISHER = {{Institute of Electrical and Electronics Engineers}}, MONTH = DEC, YEAR = {2017}, KEYWORDS = {On MPI Support ; runtime system ; sequential task flow ; task-based programming ; heterogeneous computing ; distributed computing ; multicore ; GPU ; Cholesky factorization}, PDF = {https://hal.inria.fr/hal-01618526/file/tpds14.pdf}, HAL_ID = {hal-01618526}, HAL_VERSION = {v1}, doi = {10.1109/TPDS.2017.2766064}, }
Jean Marie Couteyen Carpaye, Jean Roman, and Pierre Brenner
Design and Analysis of a Task-based Parallelization over a Runtime System of an Explicit Finite-Volume CFD Code with Adaptive Time Stepping
International Journal of Computational Science and Engineering, pp 1 - 22, 2017
[WWW] [PDF] [doi:10.1016/j.jocs.2017.03.008] Keyword(s): On Applications, HPC, CFD, runtime, task-based
@article{couteyencarpaye:hal-01507613, TITLE = {{Design and Analysis of a Task-based Parallelization over a Runtime System of an Explicit Finite-Volume CFD Code with Adaptive Time Stepping}}, AUTHOR = {Couteyen Carpaye, Jean Marie and Roman, Jean and Brenner, Pierre}, URL = {https://hal.inria.fr/hal-01507613}, JOURNAL = {{International Journal of Computational Science and Engineering}}, PUBLISHER = {{Inderscience}}, PAGES = {1 - 22}, YEAR = {2017}, DOI = {10.1016/j.jocs.2017.03.008}, KEYWORDS = {On Applications, HPC ; CFD ; runtime ; task-based}, PDF = {https://hal.inria.fr/hal-01507613/file/flusepa-task-hal-inria-preprint.pdf}, HAL_ID = {hal-01507613}, HAL_VERSION = {v1}, }
Conference articles
O. Beaumont, L. Eyraud-Dubois, and S. Kumar
Approximation Proofs of a Fast and Efficient List Scheduling Algorithm for Task-Based Runtime Systems on Multicores and GPUs
In 2017 IEEE International Parallel and Distributed Processing Symposium (IPDPS), pages 768-777, May 2017
[WWW] [PDF] [doi:10.1109/IPDPS.2017.71] Keyword(s): On Scheduling, STARPU, List scheduling, Approximation proofs, Runtime systems, Heterogeneous scheduling, Dense linear algebra
@INPROCEEDINGS{beaumont:hal-01386174, author={O. Beaumont and L. Eyraud-Dubois and S. Kumar}, booktitle={2017 IEEE International Parallel and Distributed Processing Symposium (IPDPS)}, title={Approximation Proofs of a Fast and Efficient List Scheduling Algorithm for Task-Based Runtime Systems on Multicores and GPUs}, year={2017}, volume={}, number={}, pages={768-777}, doi={10.1109/IPDPS.2017.71}, URL = {https://hal.inria.fr/hal-01386174}, PDF = {https://hal.inria.fr/hal-01386174/file/heteroPrioApproxProofsRR.pdf}, ISSN={}, month={May}, KEYWORDS = {On Scheduling; STARPU ; List scheduling ; Approximation proofs ; Runtime systems ; Heterogeneous scheduling ; Dense linear algebra}, HAL_ID = {hal-01386174}, HAL_VERSION = {v1}, }
Internal reports
Emmanuel Agullo, Bérenger Bramas, Olivier Coulaud, Luka Stanisic, and Samuel Thibault
Modeling Irregular Kernels of Task-based codes: Illustration with the Fast Multipole Method
Research Report RR-9036, INRIA Bordeaux, February 2017
[WWW] [PDF] Keyword(s): On Performance Model Tuning, Mathematical Software, Modeling and simulation, Parallel computing methodologies, fast multipole method, runtime system, task-based programming
@techreport{agullo:hal-01474556, TITLE = {{Modeling Irregular Kernels of Task-based codes: Illustration with the Fast Multipole Method}}, AUTHOR = {Agullo, Emmanuel and Bramas, B{\'e}renger and Coulaud, Olivier and Stanisic, Luka and Thibault, Samuel}, URL = {https://hal.inria.fr/hal-01474556}, TYPE = {Research Report}, NUMBER = {RR-9036}, PAGES = {35}, INSTITUTION = {{INRIA Bordeaux}}, YEAR = {2017}, MONTH = Feb, KEYWORDS = {On Performance Model Tuning; Mathematical Software ; Modeling and simulation ; Parallel computing methodologies ; fast multipole method ; runtime system ; task-based programming}, PDF = {https://hal.inria.fr/hal-01474556/file/rapport.pdf}, HAL_ID = {hal-01474556}, HAL_VERSION = {v1}, }
Emmanuel Agullo, Alfredo Buttari, Mikko Byckling, Abdou Guermouche, and Ian Masliah
Achieving high-performance with a sparse direct solver on Intel KNL
Research Report RR-9035, Inria Bordeaux Sud-Ouest ; CNRS-IRIT ; Intel corporation ; Université Bordeaux, February 2017
[WWW] [PDF] Keyword(s): On Applications, runtime system, sparse direct solver, energy efficiency, high-performance computing, portability, Intel KNL, manycore parallelism
@techreport{agullo:hal-01473475, TITLE = {{Achieving high-performance with a sparse direct solver on Intel KNL}}, AUTHOR = {Agullo, Emmanuel and Buttari, Alfredo and Byckling, Mikko and Guermouche, Abdou and Masliah, Ian}, URL = {https://hal.inria.fr/hal-01473475}, TYPE = {Research Report}, NUMBER = {RR-9035}, PAGES = {15}, INSTITUTION = {{Inria Bordeaux Sud-Ouest ; CNRS-IRIT ; Intel corporation ; Universit{\'e} Bordeaux}}, YEAR = {2017}, MONTH = Feb, KEYWORDS = {On Applications; runtime system ; sparse direct solver ; energy efficiency ; high-performance computing ; portability ; Intel KNL ; manycore parallelism}, PDF = {https://hal.inria.fr/hal-01473475/file/RR-9035.pdf}, HAL_ID = {hal-01473475}, HAL_VERSION = {v1}, }