Details
| Originalsprache | Englisch |
|---|---|
| Titel des Sammelwerks | 14th International Symposium on Reconfigurable Communication-centric Systems-on-Chip (ReCoSoC 2019) |
| Untertitel | Proceedings |
| Herausgeber/-innen | William Fornaciari, David Novo, Leandro Soares Indrusiak |
| Herausgeber (Verlag) | Institute of Electrical and Electronics Engineers Inc. |
| Seiten | 74-81 |
| Seitenumfang | 8 |
| ISBN (elektronisch) | 978-1-7281-4770-3 |
| ISBN (Print) | 978-1-7281-4771-0 |
| Publikationsstatus | Veröffentlicht - Juli 2019 |
| Veranstaltung | 14th International Symposium on Reconfigurable Communication-centric Systems-on-Chip - York, Großbritannien / Vereinigtes Königreich Dauer: 1 Juli 2019 → 3 Juli 2019 |
Publikationsreihe
| Name | International Symposium on Reconfigurable and Communication-Centric Systems-on-Chip (ReCoSoC) |
|---|---|
| ISSN (Print) | 2642-7230 |
| ISSN (elektronisch) | 2642-7222 |
Abstract
In NoC-based multi-core systems, the communication subsystem accounts for a significant fraction of the energy consumption. Traditional NoCs guarantee error-free data transfer at the expenses of limiting the hardware optimization possibilities. However, in various applications, such as multimedia processing and machine learning, inaccurate outputs within an acceptable variance is tolerable. Recently, approximate NOCs have been proposed as a viable solution to exploit the trade off between the energy efficiency with data approximation. In this context, we propose two coding techniques. First, a combined memory-less integer-value encoding is proposed to reduce the integer-value deviation in the body flit of the approximate NOCs; and second, a crosstalk avoidance code (CAC) to guarantees the exact transmission of the head flit. Our technique improves the energy consumption of the approximate NOCs remarkably and does not introduces any performance or area penalty. Experimental results using a 65nm commercial technology asses the advantages of the proposed coding approach versus conventional transmission. proposed coding in real scenarios.
ASJC Scopus Sachgebiete
- Informatik (insg.)
- Software
- Informatik (insg.)
- Hardware und Architektur
- Informatik (insg.)
- Computernetzwerke und -kommunikation
Ziele für nachhaltige Entwicklung
Zitieren
- Standard
- Harvard
- Apa
- Vancouver
- BibTex
- RIS
14th International Symposium on Reconfigurable Communication-centric Systems-on-Chip (ReCoSoC 2019): Proceedings. Hrsg. / William Fornaciari; David Novo; Leandro Soares Indrusiak. Institute of Electrical and Electronics Engineers Inc., 2019. S. 74-81 9034965 (International Symposium on Reconfigurable and Communication-Centric Systems-on-Chip (ReCoSoC)).
Publikation: Beitrag in Buch/Bericht/Sammelwerk/Konferenzband › Aufsatz in Konferenzband › Forschung › Peer-Review
}
TY - GEN
T1 - A Coding Approach to Improve the Energy Efficiency of Approximate NoCs
AU - Najafi, Amir
AU - Bamberg, Lennart
AU - Payá Vayá, Guillermo
AU - Ortiz, Alberto García
N1 - Funding information: This work is funded by the German Research Foundation (DFG) project GA 763/4-1.
PY - 2019/7
Y1 - 2019/7
N2 - In NoC-based multi-core systems, the communication subsystem accounts for a significant fraction of the energy consumption. Traditional NoCs guarantee error-free data transfer at the expenses of limiting the hardware optimization possibilities. However, in various applications, such as multimedia processing and machine learning, inaccurate outputs within an acceptable variance is tolerable. Recently, approximate NOCs have been proposed as a viable solution to exploit the trade off between the energy efficiency with data approximation. In this context, we propose two coding techniques. First, a combined memory-less integer-value encoding is proposed to reduce the integer-value deviation in the body flit of the approximate NOCs; and second, a crosstalk avoidance code (CAC) to guarantees the exact transmission of the head flit. Our technique improves the energy consumption of the approximate NOCs remarkably and does not introduces any performance or area penalty. Experimental results using a 65nm commercial technology asses the advantages of the proposed coding approach versus conventional transmission. proposed coding in real scenarios.
AB - In NoC-based multi-core systems, the communication subsystem accounts for a significant fraction of the energy consumption. Traditional NoCs guarantee error-free data transfer at the expenses of limiting the hardware optimization possibilities. However, in various applications, such as multimedia processing and machine learning, inaccurate outputs within an acceptable variance is tolerable. Recently, approximate NOCs have been proposed as a viable solution to exploit the trade off between the energy efficiency with data approximation. In this context, we propose two coding techniques. First, a combined memory-less integer-value encoding is proposed to reduce the integer-value deviation in the body flit of the approximate NOCs; and second, a crosstalk avoidance code (CAC) to guarantees the exact transmission of the head flit. Our technique improves the energy consumption of the approximate NOCs remarkably and does not introduces any performance or area penalty. Experimental results using a 65nm commercial technology asses the advantages of the proposed coding approach versus conventional transmission. proposed coding in real scenarios.
UR - http://www.scopus.com/inward/record.url?scp=85084404505&partnerID=8YFLogxK
UR - https://dblp.org/rec/conf/recosoc/NajafiBVO19
U2 - 10.1109/RECOSOC48741.2019.9034965
DO - 10.1109/RECOSOC48741.2019.9034965
M3 - Conference contribution
SN - 978-1-7281-4771-0
T3 - International Symposium on Reconfigurable and Communication-Centric Systems-on-Chip (ReCoSoC)
SP - 74
EP - 81
BT - 14th International Symposium on Reconfigurable Communication-centric Systems-on-Chip (ReCoSoC 2019)
A2 - Fornaciari, William
A2 - Novo, David
A2 - Indrusiak, Leandro Soares
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 14th International Symposium on Reconfigurable Communication-centric Systems-on-Chip
Y2 - 1 July 2019 through 3 July 2019
ER -