Carbon-Efficient 3D DNN Acceleration: Optimizing Performance and Sustainability
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866916765958668288 |
|---|---|
| author | Panteleaki, Aikaterini Maria Balaskas, Konstantinos Zervakis, Georgios Amrouch, Hussam Anagnostopoulos, Iraklis |
| author_facet | Panteleaki, Aikaterini Maria Balaskas, Konstantinos Zervakis, Georgios Amrouch, Hussam Anagnostopoulos, Iraklis |
| contents | As Deep Neural Networks (DNNs) continue to drive advancements in artificial intelligence, the design of hardware accelerators faces growing concerns over embodied carbon footprint due to complex fabrication processes. 3D integration improves performance but introduces sustainability challenges, making carbon-aware optimization essential. In this work, we propose a carbon-efficient design methodology for 3D DNN accelerators, leveraging approximate computing and genetic algorithm-based design space exploration to optimize Carbon Delay Product (CDP). By integrating area-efficient approximate multipliers into Multiply-Accumulate (MAC) units, our approach effectively reduces silicon area and fabrication overhead while maintaining high computational accuracy. Experimental evaluations across three technology nodes (45nm, 14nm, and 7nm) show that our method reduces embodied carbon by up to 30% with negligible accuracy drop. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_09851 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Carbon-Efficient 3D DNN Acceleration: Optimizing Performance and Sustainability Panteleaki, Aikaterini Maria Balaskas, Konstantinos Zervakis, Georgios Amrouch, Hussam Anagnostopoulos, Iraklis Hardware Architecture Artificial Intelligence As Deep Neural Networks (DNNs) continue to drive advancements in artificial intelligence, the design of hardware accelerators faces growing concerns over embodied carbon footprint due to complex fabrication processes. 3D integration improves performance but introduces sustainability challenges, making carbon-aware optimization essential. In this work, we propose a carbon-efficient design methodology for 3D DNN accelerators, leveraging approximate computing and genetic algorithm-based design space exploration to optimize Carbon Delay Product (CDP). By integrating area-efficient approximate multipliers into Multiply-Accumulate (MAC) units, our approach effectively reduces silicon area and fabrication overhead while maintaining high computational accuracy. Experimental evaluations across three technology nodes (45nm, 14nm, and 7nm) show that our method reduces embodied carbon by up to 30% with negligible accuracy drop. |
| title | Carbon-Efficient 3D DNN Acceleration: Optimizing Performance and Sustainability |
| topic | Hardware Architecture Artificial Intelligence |
| url | https://arxiv.org/abs/2504.09851 |