Home > Publications database > Efficient Distributed GPU Programming for Exascale > print

001     903614
005     20250822121434.0
024 7 _ |a 10.5281/ZENODO.5745505
|2 doi
024 7 _ |a altmetric:117880295
|2 altmetric
037 _ _ |a FZJ-2021-05268
041 _ _ |a English
100 1 _ |a Garcia de Gonzalo, Simon
|0 P:(DE-HGF)0
|b 0
111 2 _ |a Supercomputing Conference 2021
|c online
|d 2021-11-14 - 2021-11-14
|g SC21
245 _ _ |a Efficient Distributed GPU Programming for Exascale
260 _ _ |c 2021
336 7 _ |a lecture
|2 DRIVER
336 7 _ |a Generic
|0 31
|2 EndNote
336 7 _ |a MISC
|2 BibTeX
336 7 _ |a Lecture
|b lecture
|m lecture
|0 PUB:(DE-HGF)17
|s 1639660672_30760
|2 PUB:(DE-HGF)
|x After Call
336 7 _ |a LECTURE_SPEECH
|2 ORCID
336 7 _ |a Text
|2 DataCite
500 _ _ |a Tutorial at SC21 Conference, consisting of lectures and hands-on exercises.
520 _ _ |a Over the past years, GPUs became ubiquitous in HPC installations around the world. Today, they provide the majority of performance of some of the largest supercomputers (e.g. Summit, Sierra, JUWELS Booster). This trend continues in upcoming pre-exascale and exascale systems (LUMI, Leonardo; Frontier): GPUs are chosen as the core computing devices to enter this next era of HPC.To take advantage of future GPU-accelerated systems with tens of thousands of devices, application developers need to have the proper skills and tools to understand, manage, and optimize distributed GPU applications.

In this tutorial, participants will learn techniques to efficiently program large-scale multi-GPU systems. While programming multiple GPUs with MPI is explained in detail, also advanced techniques and models (NCCL, NVSHMEM, …) are presented. Tools for analysis are used to motivate implementation of performance optimizations. The tutorial combines lectures and hands-on exercises, using Europe's fastest supercomputer, JUWELS Booster with NVIDIA A100 GPUs.
536 _ _ |a 5121 - Supercomputing & Big Data Facilities (POF4-512)
|0 G:(DE-HGF)POF4-5121
|c POF4-512
|f POF IV
|x 0
536 _ _ |a 5111 - Domain-Specific Simulation & Data Life Cycle Labs (SDLs) and Research Groups (POF4-511)
|0 G:(DE-HGF)POF4-5111
|c POF4-511
|f POF IV
|x 1
536 _ _ |a ATML-X-DEV - ATML Accelerating Devices (ATML-X-DEV)
|0 G:(DE-Juel-1)ATML-X-DEV
|c ATML-X-DEV
|x 2
588 _ _ |a Dataset connected to DataCite
700 1 _ |a Hrywniak, Markus
|0 P:(DE-Juel1)180799
|b 1
|u fzj
700 1 _ |a Kraus, Jiri
|0 P:(DE-Juel1)137023
|b 2
|u fzj
700 1 _ |a Oden, Lena
|0 P:(DE-Juel1)188270
|b 3
|u fzj
700 1 _ |a Herten, Andreas
|0 P:(DE-Juel1)145478
|b 4
|e Corresponding author
|u fzj
773 _ _ |a 10.5281/ZENODO.5745505
909 C O |o oai:juser.fz-juelich.de:903614
|p VDB
910 1 _ |a Forschungszentrum Jülich
|0 I:(DE-588b)5008462-8
|k FZJ
|b 1
|6 P:(DE-Juel1)180799
910 1 _ |a Forschungszentrum Jülich
|0 I:(DE-588b)5008462-8
|k FZJ
|b 2
|6 P:(DE-Juel1)137023
910 1 _ |a Forschungszentrum Jülich
|0 I:(DE-588b)5008462-8
|k FZJ
|b 3
|6 P:(DE-Juel1)188270
910 1 _ |a Forschungszentrum Jülich
|0 I:(DE-588b)5008462-8
|k FZJ
|b 4
|6 P:(DE-Juel1)145478
913 1 _ |a DE-HGF
|b Key Technologies
|l Engineering Digital Futures – Supercomputing, Data Management and Information Security for Knowledge and Action
|1 G:(DE-HGF)POF4-510
|0 G:(DE-HGF)POF4-512
|3 G:(DE-HGF)POF4
|2 G:(DE-HGF)POF4-500
|4 G:(DE-HGF)POF
|v Supercomputing & Big Data Infrastructures
|9 G:(DE-HGF)POF4-5121
|x 0
913 1 _ |a DE-HGF
|b Key Technologies
|l Engineering Digital Futures – Supercomputing, Data Management and Information Security for Knowledge and Action
|1 G:(DE-HGF)POF4-510
|0 G:(DE-HGF)POF4-511
|3 G:(DE-HGF)POF4
|2 G:(DE-HGF)POF4-500
|4 G:(DE-HGF)POF
|v Enabling Computational- & Data-Intensive Science and Engineering
|9 G:(DE-HGF)POF4-5111
|x 1
914 1 _ |y 2021
920 1 _ |0 I:(DE-Juel1)JSC-20090406
|k JSC
|l Jülich Supercomputing Center
|x 0
980 _ _ |a lecture
980 _ _ |a VDB
980 _ _ |a I:(DE-Juel1)JSC-20090406
980 _ _ |a UNRESTRICTED

LibraryCollectionCLSMajorCLSMinorLanguageAuthor
Marc 21