elib
DLR-Header
DLR-Logo -> http://www.dlr.de
DLR Portal Home | Impressum | Datenschutz | Kontakt | English
Schriftgröße: [-] Text [+]

Early Timing and Energy Prediction and Optimization of Artificial Neural Networks on Multi-Core Platforms

Dariol, Quentin (2023) Early Timing and Energy Prediction and Optimization of Artificial Neural Networks on Multi-Core Platforms. Dissertation, Nantes Université. doi: 10.5281/zenodo.11207819.

[img] PDF
28MB
[img] PDF
5MB

Offizielle URL: https://theses.hal.science/tel-04390337v1

Kurzfassung

The need to implement artificial Neural Networks (NNs) on embedded multicore platforms has become fundamental. Predicting timing properties (inference time, latency, throughput) and energy as early as possible in the design process is necessary to find solutions that optimize resource use and respect the constraints imposed on the system. A major modeling difficulty comes from the need to correctly describe the influence of resource sharing (processor, memory, communication bus) within multi-core platforms. In this thesis, we present a complete flow for predicting and optimizing timing properties and energy, combining several modeling approaches. This flow leads to optimized resource occupancy without degrading the performance of implemented NNs. Predictions are compared with measurements on real targets. The proposed models have an accuracy of over 97% on timing and 93% on energy for 54 mappings of 4 NNs, with a prediction time of 20s per mapping. We show how to use the models to efficiently explore the design space and find optimized solutions that satisfy the constraints imposed on the system.

elib-URL des Eintrags:https://elib.dlr.de/200347/
Dokumentart:Hochschulschrift (Dissertation)
Titel:Early Timing and Energy Prediction and Optimization of Artificial Neural Networks on Multi-Core Platforms
Autoren:
AutorenInstitution oder E-Mail-AdresseAutoren-ORCID-iDORCID Put Code
Dariol, Quentinquentin.dariol (at) dlr.dehttps://orcid.org/0000-0002-3284-6882NICHT SPEZIFIZIERT
Datum:2023
Referierte Publikation:Nein
Open Access:Ja
DOI:10.5281/zenodo.11207819
Seitenanzahl:188
Status:veröffentlicht
Stichwörter:Embedded artificial intelligence, system level design, timing and energy prediction
Institution:Nantes Université
Abteilung:Électronique
HGF - Forschungsbereich:Luftfahrt, Raumfahrt und Verkehr
HGF - Programm:Verkehr
HGF - Programmthema:keine Zuordnung
DLR - Schwerpunkt:Verkehr
DLR - Forschungsgebiet:V - keine Zuordnung
DLR - Teilgebiet (Projekt, Vorhaben):V - keine Zuordnung
Standort: Oldenburg
Institute & Einrichtungen:Institut für Systems Engineering für zukünftige Mobilität > System Evolution and Operation
Hinterlegt von: Dariol, Quentin
Hinterlegt am:08 Jan 2024 07:56
Letzte Änderung:21 Mai 2024 07:48

Nur für Mitarbeiter des Archivs: Kontrollseite des Eintrags

Blättern
Suchen
Hilfe & Kontakt
Informationen
electronic library verwendet EPrints 3.3.12
Gestaltung Webseite und Datenbank: Copyright © Deutsches Zentrum für Luft- und Raumfahrt (DLR). Alle Rechte vorbehalten.