Sommer, Lukas ; Stock, Florian-Wolfgang ; Solis-Vasquez, Leonardo ; Koch, Andreas
Hrsg.: Forschungsvereinigung Automobieltechnik (2019)
EPHoS: Evaluation of Programming - Models for Heterogeneous Systems.
Report, Bibliographie
Kurzbeschreibung (Abstract)
In recent years, there has been a steep increase in the computational demands of automotive applications, in particular in the field of assisted and autonomous driving. With the slowing down of Moore’s Law, these computational requirements can no longer be satisfied by classical, embedded single-core processors. Therefore, automotive manufacturers, suppliers, and original equipment manufacturers (OEM) turn towards parallel and heterogeneous platforms, combining multi-core CPUs and dedicated accelerators on a single board or SoC to satisfy their application’s compute demands. As the programming languages traditionally used in this industry, specifically C and C++, do not provide abstraction mechanisms to leverage the parallelism in such heterogeneous platforms, new, parallel programming models need to be considered for adoption in the industry. In this work, we evaluate the applicability of parallel programming mod- els established, e.g., in high-performance computing, for embedded use in the automotive domain, targeting parallel and heterogeneous embedded computing platforms. To this end, we conduct a practical implementation survey with benchmarks from the autonomous driving domain, targeting representative execution platforms, and evaluate important usability aspects of the investigated programming models.
| Typ des Eintrags: | Report |
|---|---|
| Erschienen: | 2019 |
| Autor(en): | Sommer, Lukas ; Stock, Florian-Wolfgang ; Solis-Vasquez, Leonardo ; Koch, Andreas |
| Art des Eintrags: | Bibliographie |
| Titel: | EPHoS: Evaluation of Programming - Models for Heterogeneous Systems |
| Sprache: | Englisch |
| Publikationsjahr: | 17 Juni 2019 |
| Ort: | Berlin |
| Verlag: | Forschungsvereinigung Automobiltechnik |
| Reihe: | FAT-Schriftenreihe |
| Band einer Reihe: | 317 |
| URL / URN: | https://www.tib.eu/de/suchen/id/TIBKAT:166997393X/EPHoS-Eval... |
| Kurzbeschreibung (Abstract): | In recent years, there has been a steep increase in the computational demands of automotive applications, in particular in the field of assisted and autonomous driving. With the slowing down of Moore’s Law, these computational requirements can no longer be satisfied by classical, embedded single-core processors. Therefore, automotive manufacturers, suppliers, and original equipment manufacturers (OEM) turn towards parallel and heterogeneous platforms, combining multi-core CPUs and dedicated accelerators on a single board or SoC to satisfy their application’s compute demands. As the programming languages traditionally used in this industry, specifically C and C++, do not provide abstraction mechanisms to leverage the parallelism in such heterogeneous platforms, new, parallel programming models need to be considered for adoption in the industry. In this work, we evaluate the applicability of parallel programming mod- els established, e.g., in high-performance computing, for embedded use in the automotive domain, targeting parallel and heterogeneous embedded computing platforms. To this end, we conduct a practical implementation survey with benchmarks from the autonomous driving domain, targeting representative execution platforms, and evaluate important usability aspects of the investigated programming models. |
| Fachbereich(e)/-gebiet(e): | 20 Fachbereich Informatik 20 Fachbereich Informatik > Eingebettete Systeme und ihre Anwendungen |
| Hinterlegungsdatum: | 22 Dez 2020 13:08 |
| Letzte Änderung: | 10 Aug 2023 14:11 |
| PPN: | |
| Export: | |
| Suche nach Titel in: | TUfind oder in Google |
 |
Frage zum Eintrag |
Optionen (nur für Redakteure)
 |
Redaktionelle Details anzeigen |
Drucken
Impressum