Modeling and Investigating the LLM electricity demand's impacts on Power and Energy Systems
This is the repo for The Unseen AI Disruptions for Power Grids: LLM-Induced Transients
Authors: Yuzhuo Li, Mariam Mughees, Yize Chen, Yunwei Ryan Li
University of Alberta
Contact: [email protected]
Recent breakthroughs of large language models (LLMs) have exhibited superior capability across major industries and stimulated multi-hundred-billion-dollar investment in AI-centric data centers in the next 3-5 years. This, in turn, bring the increasing concerns on sustainability and AI-related power and energy usage. However, there is a largely overlooked issue as challenging and critical as AI model and infrastructure efficiency: the disruptive dynamic power consumption behaviour. With fast, transient dynamics, AI infrastructure features ultra-low inertia, sharp power surge and dip, and a significant peak-idle power ratio. We develop high-level mathematical models to depict AI workload behaviour and discuss the multifaceted challenges and opportunities they potentially bring to existing power grids. Observing the rapidly evolving machine learning (ML) and AI technologies, this project emphasizes the critical need for interdisciplinary approaches to ensure reliable and sustainable AI infrastructure development, and provides a starting point for researchers and practitioners to tackle such challenges.
Consider the sizes of the models, the size of the data, along with the hardware limits on memory and computation, we are curating a list of results from experiments with feasible configurations.
Links to Key Repositories:
Links to the paper:
- [Our paper] https://arxiv.org/abs/2409.11416
- [To cite] Yuzhuo Li, Mariam Mughees, Yize Chen, and Yunwei Ryan Li. "The Unseen AI Disruptions for Power Grids: LLM-Induced Transients." arXiv preprint arXiv:2409.11416 (2024).
