Beyond Efficiency: Scaling AI Sustainably

TL;DR

This paper examines the carbon footprint of AI across its entire lifecycle, emphasizing the need for sustainable scaling by optimizing efficiency from hardware manufacturing to datacenter operations.

Contribution

It provides a comprehensive analysis of AI's carbon impact and highlights opportunities for efficiency improvements throughout the AI development and deployment cycle.

Findings

Characterizes operational and embodied carbon emissions of AI

Identifies key efficiency optimization opportunities in AI technologies

Advocates for lifecycle-wide sustainability strategies

Abstract

Barroso's seminal contributions in energy-proportional warehouse-scale computing launched an era where modern datacenters have become more energy efficient and cost effective than ever before. At the same time, modern AI applications have driven ever-increasing demands in computing, highlighting the importance of optimizing efficiency across the entire deep learning model development cycle. This paper characterizes the carbon impact of AI, including both operational carbon emissions from training and inference as well as embodied carbon emissions from datacenter construction and hardware manufacturing. We highlight key efficiency optimization opportunities for cutting-edge AI technologies, from deep learning recommendation models to multi-modal generative AI tasks. To scale AI sustainably, we must also go beyond efficiency and optimize across the life cycle of computing infrastructures, from hardware manufacturing to datacenter operations and end-of-life processing for the hardware.