Generating and Exploiting Deep Learning Variants to Increase Utilization of the Heterogeneous Resources in Autonomous Driving Platforms

Nowadays, Deep learning-based solutions and, in particular, deep neural networks (DNNs) are getting into several core functionalities in critical real-time embedded systems (CRTES), like those in planes, cars, and satellites, from vision-based perception (object detection and object tracking) system...

Descripción completa

Detalles Bibliográficos
Autor: Pujol Torramorell, Roger
Tipo de recurso: tesis de maestría
Fecha de publicación:2020
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/334558
Acceso en línea:https://hdl.handle.net/2117/334558
Access Level:acceso abierto
Palabra clave:Neural networks (Computer science)
Real-time data processing
conducció autònoma
xarxes neuronals profundes
programació lineal d'enters
recursos heterogenis
planificació
sistemes integrats crítics en temps real
autonomous driving
AD
DNN
integer linear programming
ILP
GPU
heterogeneous resources
scheduling
critical real-time embedded systems
CRTES
real-time
Xarxes neuronals (Informàtica)
Temps real (Informàtica)
Àrees temàtiques de la UPC::Informàtica
id ES_03314f252ea9d464c5d25faae4de233a
oai_identifier_str oai:upcommons.upc.edu:2117/334558
network_acronym_str ES
network_name_str España
repository_id_str
spelling Generating and Exploiting Deep Learning Variants to Increase Utilization of the Heterogeneous Resources in Autonomous Driving PlatformsPujol Torramorell, RogerNeural networks (Computer science)Real-time data processingconducció autònomaxarxes neuronals profundesprogramació lineal d'entersrecursos heterogenisplanificaciósistemes integrats crítics en temps realautonomous drivingADDNNinteger linear programmingILPGPUheterogeneous resourcesschedulingcritical real-time embedded systemsCRTESreal-timeXarxes neuronals (Informàtica)Temps real (Informàtica)Àrees temàtiques de la UPC::InformàticaNowadays, Deep learning-based solutions and, in particular, deep neural networks (DNNs) are getting into several core functionalities in critical real-time embedded systems (CRTES), like those in planes, cars, and satellites, from vision-based perception (object detection and object tracking) systems to trajectory planning. As a result, several deep learning instances are running simultaneously at any time on the same computing platform. However, while modern computing platforms offer a variety of computing elements (e.g., CPUs, GPUs, and specific accelerators) in which those DNN instances can be executed depending on their computational requirements and temporal constraints. Currently, most DNNs are mainly programmed to exploit one particular computing element, regular cores of the GPUs. This lack of variety causes a resource imbalance and under-utilization of the various computing element resources when executing several DNN instances, causing an increase in DNN tasks' execution time requirements. In this Thesis, (a) we develop different variants (implementation) of well-known DNN libraries used in the Apollo Autonomous Driving software for each of the computing elements of the latest NVIDIA Xavier system-on-chip. Each variant is configured to balance resource requirements and performance: the regular CPU core implementation that can run on 2, 4, and 6 cores (always leaving 2 cores free for other computations); the GPU with regular and Tensor cores variants that can run on 4 or 8 GPU's Stream Multiprocessors (SM); and 1 or 2 NVIDIA's Deep Learning Accelerators (NVDLA); (b) we show that each particular variant/configuration offers different resource utilization/performance point. (c) we show how those heterogeneous computing elements can be exploited by a static scheduler to sustain the execution of multiple and diverse DNN variants on the same platform.Universitat Politècnica de CatalunyaHamid TabaniKosmidis, Leonidas20202020-06-2420202020-12-16master thesishttp://purl.org/coar/resource_type/c_bdccNAhttp://purl.org/coar/version/c_be7fb7dd8ff6fe43info:eu-repo/semantics/masterThesisapplication/pdfhttps://hdl.handle.net/2117/334558reponame:UPCommons. Portal del coneixement obert de la UPCinstname:Universitat Politècnica de Catalunya (UPC)Inglésengopen accesshttp://purl.org/coar/access_right/c_abf2info:eu-repo/semantics/openAccessoai:upcommons.upc.edu:2117/3345582026-05-27T15:37:01Z
dc.title.none.fl_str_mv Generating and Exploiting Deep Learning Variants to Increase Utilization of the Heterogeneous Resources in Autonomous Driving Platforms
title Generating and Exploiting Deep Learning Variants to Increase Utilization of the Heterogeneous Resources in Autonomous Driving Platforms
spellingShingle Generating and Exploiting Deep Learning Variants to Increase Utilization of the Heterogeneous Resources in Autonomous Driving Platforms
Pujol Torramorell, Roger
Neural networks (Computer science)
Real-time data processing
conducció autònoma
xarxes neuronals profundes
programació lineal d'enters
recursos heterogenis
planificació
sistemes integrats crítics en temps real
autonomous driving
AD
DNN
integer linear programming
ILP
GPU
heterogeneous resources
scheduling
critical real-time embedded systems
CRTES
real-time
Xarxes neuronals (Informàtica)
Temps real (Informàtica)
Àrees temàtiques de la UPC::Informàtica
title_short Generating and Exploiting Deep Learning Variants to Increase Utilization of the Heterogeneous Resources in Autonomous Driving Platforms
title_full Generating and Exploiting Deep Learning Variants to Increase Utilization of the Heterogeneous Resources in Autonomous Driving Platforms
title_fullStr Generating and Exploiting Deep Learning Variants to Increase Utilization of the Heterogeneous Resources in Autonomous Driving Platforms
title_full_unstemmed Generating and Exploiting Deep Learning Variants to Increase Utilization of the Heterogeneous Resources in Autonomous Driving Platforms
title_sort Generating and Exploiting Deep Learning Variants to Increase Utilization of the Heterogeneous Resources in Autonomous Driving Platforms
dc.creator.none.fl_str_mv Pujol Torramorell, Roger
author Pujol Torramorell, Roger
author_facet Pujol Torramorell, Roger
author_role author
dc.contributor.none.fl_str_mv Hamid Tabani
Kosmidis, Leonidas
dc.subject.none.fl_str_mv Neural networks (Computer science)
Real-time data processing
conducció autònoma
xarxes neuronals profundes
programació lineal d'enters
recursos heterogenis
planificació
sistemes integrats crítics en temps real
autonomous driving
AD
DNN
integer linear programming
ILP
GPU
heterogeneous resources
scheduling
critical real-time embedded systems
CRTES
real-time
Xarxes neuronals (Informàtica)
Temps real (Informàtica)
Àrees temàtiques de la UPC::Informàtica
topic Neural networks (Computer science)
Real-time data processing
conducció autònoma
xarxes neuronals profundes
programació lineal d'enters
recursos heterogenis
planificació
sistemes integrats crítics en temps real
autonomous driving
AD
DNN
integer linear programming
ILP
GPU
heterogeneous resources
scheduling
critical real-time embedded systems
CRTES
real-time
Xarxes neuronals (Informàtica)
Temps real (Informàtica)
Àrees temàtiques de la UPC::Informàtica
description Nowadays, Deep learning-based solutions and, in particular, deep neural networks (DNNs) are getting into several core functionalities in critical real-time embedded systems (CRTES), like those in planes, cars, and satellites, from vision-based perception (object detection and object tracking) systems to trajectory planning. As a result, several deep learning instances are running simultaneously at any time on the same computing platform. However, while modern computing platforms offer a variety of computing elements (e.g., CPUs, GPUs, and specific accelerators) in which those DNN instances can be executed depending on their computational requirements and temporal constraints. Currently, most DNNs are mainly programmed to exploit one particular computing element, regular cores of the GPUs. This lack of variety causes a resource imbalance and under-utilization of the various computing element resources when executing several DNN instances, causing an increase in DNN tasks' execution time requirements. In this Thesis, (a) we develop different variants (implementation) of well-known DNN libraries used in the Apollo Autonomous Driving software for each of the computing elements of the latest NVIDIA Xavier system-on-chip. Each variant is configured to balance resource requirements and performance: the regular CPU core implementation that can run on 2, 4, and 6 cores (always leaving 2 cores free for other computations); the GPU with regular and Tensor cores variants that can run on 4 or 8 GPU's Stream Multiprocessors (SM); and 1 or 2 NVIDIA's Deep Learning Accelerators (NVDLA); (b) we show that each particular variant/configuration offers different resource utilization/performance point. (c) we show how those heterogeneous computing elements can be exploited by a static scheduler to sustain the execution of multiple and diverse DNN variants on the same platform.
publishDate 2020
dc.date.none.fl_str_mv 2020
2020-06-24
2020
2020-12-16
dc.type.none.fl_str_mv master thesis
http://purl.org/coar/resource_type/c_bdcc
NA
http://purl.org/coar/version/c_be7fb7dd8ff6fe43
dc.type.openaire.fl_str_mv info:eu-repo/semantics/masterThesis
format masterThesis
dc.identifier.none.fl_str_mv https://hdl.handle.net/2117/334558
url https://hdl.handle.net/2117/334558
dc.language.none.fl_str_mv Inglés
eng
language_invalid_str_mv Inglés
language eng
dc.rights.none.fl_str_mv open access
http://purl.org/coar/access_right/c_abf2
dc.rights.openaire.fl_str_mv info:eu-repo/semantics/openAccess
rights_invalid_str_mv open access
http://purl.org/coar/access_right/c_abf2
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Universitat Politècnica de Catalunya
publisher.none.fl_str_mv Universitat Politècnica de Catalunya
dc.source.none.fl_str_mv reponame:UPCommons. Portal del coneixement obert de la UPC
instname:Universitat Politècnica de Catalunya (UPC)
instname_str Universitat Politècnica de Catalunya (UPC)
reponame_str UPCommons. Portal del coneixement obert de la UPC
collection UPCommons. Portal del coneixement obert de la UPC
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869402708697415680
score 15,301603

Similares en LA Referencia