Carbon-Efficient 3D DNN Acceleration: Optimizing Performance and Sustainability
This addresses sustainability challenges in hardware design for AI, offering a domain-specific incremental improvement.
The paper tackles the problem of high embodied carbon footprint in 3D DNN accelerators by proposing a carbon-efficient design methodology, achieving up to 30% reduction in embodied carbon with negligible accuracy drop.
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.