dc.contributor.author |
Toledo Diaz, Leonel Antonio |
dc.contributor.author |
Valero-Lara, Pedro |
dc.contributor.author |
Vetter, Jeffrey |
dc.contributor.author |
Peña, Antonio J. |
dc.date.accessioned |
2023-02-23T08:49:57Z |
dc.date.available |
2023-12-01T01:45:07Z |
dc.date.issued |
2021-12-01 |
dc.identifier.uri |
http://hdl.handle.net/2072/531315 |
dc.format.extent |
6 p. |
dc.language.iso |
eng |
dc.publisher |
IEEE |
dc.relation.ispartof |
2021 IEEE 28th International Conference on High Performance Computing, Data, and Analytics (HiPC) |
dc.rights |
© 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. |
dc.source |
RECERCAT (Dipòsit de la Recerca de Catalunya) |
dc.subject.other |
Xarxes d'àrea extensa (Ordinadors) |
dc.subject.other |
Virtual Reality |
dc.subject.other |
GPU Programming |
dc.subject.other |
Computer Graphics |
dc.subject.other |
Media Internet Area |
dc.title |
Static Graphs for Coding Productivity in OpenACC |
dc.type |
info:eu-repo/semantics/article |
dc.type |
info:eu-repo/semantics/acceptedVersion |
dc.subject.udc |
621.3 - Enginyeria elèctrica. Electrotècnia. Telecomunicacions |
dc.embargo.terms |
24 mesos |
dc.identifier.doi |
10.1109/HiPC53243.2021.00050 |
dc.rights.accessLevel |
info:eu-repo/semantics/openAccess |
dc.description.abstract |
The main contribution of this work is to increase the coding productivity for GPU programming by using the concept of Static Graphs. To do so, we have combined the new CUDA Graph API with the OpenACC programming model. We use as test cases a well-known and widely used problems in HPC and AI: the Particle Swarm Optimization. We complement the OpenACC functionality with the use of CUDA Graph, achieving accelerations of more than one order of magnitude, and a performance very close to a reference and optimized CUDA code. Finally, we propose a new specification to incorporate the concept of Static Graphs into the OpenACC specification. |