2025 · 2025 IEEE International Conference on Collaborative Advances in Software and COmputiNg (CASCON)
Energy Consumption Analysis of Large Language Models Across CPU and GPU Using Diverse Metric Types
Evidence basis: abstract-and-metadata-reviewed · Review status: catalog-reviewed; paper-author approval pending
energy-efficiency llm-efficiency performance-modeling machine-learning performance-analysis
LLM energy consumption CPU energy GPU energy green AI static metrics dynamic metrics NLP workloads energy prediction CASCON 2025
Core contribution: The abstract-level contribution is a broader CPU/GPU energy-analysis and prediction perspective for LLM workloads that combines multiple metric types and task coverage.
Catalog abstract summary
The paper presents a multi-level analysis of energy consumption for large language model workloads across CPU and GPU settings, combining diverse static and dynamic metrics to build and validate energy-prediction models across NLP tasks.
Source: Public abstract/metadata mirror and IEEE bibliographic record, paraphrased; full text not obtained.
Problem and motivation
Resource-utilization measurements alone may not capture the energy behavior of LLM workloads, while existing prediction models may not cover diverse NLP tasks and metric types.
Method and contribution
The accessible abstract describes collecting and correlating dynamic and static metrics across LLM/NLP models and tasks, then validating energy-prediction models. The exact tracer or power instrument, hardware, model names, metric definitions, equations, and validation protocol require the full paper.
Findings and evidence
The public abstract claims robust energy prediction across diverse NLP tasks. No exact error, correlation, energy value, workload count, CPU/GPU configuration, or baseline result was verified.
Limitations and future directions
Limitations: Only metadata and abstract-level evidence were available; the implementation, datasets, workloads, quantitative evaluation, threats to validity, and paper-specific limitations remain unverified.
Future work: The paper-specific future-work section remains unverified and should be added only after a legal full-text copy is read.
Sources and identifiers
- Published version published
- IEEE record and public abstract metadata public_abstract
When to cite this paper
Cite this paper when its specific method, evidence, or benchmark is directly relevant.
- The paper's method is directly relevant.
- The paper's evidence or benchmark is directly relevant.
Citation
@inproceedings{ezzatiJivan2025energyconsumption,
author = {Tong Zhang and Leila Tahmooresnejad and Naser Ezzati-Jivan},
title = {Energy Consumption Analysis of Large Language Models Across CPU and GPU Using Diverse Metric Types},
year = {2025},
booktitle = {2025 IEEE International Conference on Collaborative Advances in Software and COmputiNg (CASCON)},
pages = {301-310},
publisher = {IEEE},
doi = {10.1109/CASCON66301.2025.00056},
url = {https://doi.org/10.1109/CASCON66301.2025.00056}
}Other citation formats for Word and reference managers
Zhang, T., Tahmooresnejad, L., & Ezzati-Jivan, N. (2025). Energy Consumption Analysis of Large Language Models Across CPU and GPU Using Diverse Metric Types. In 2025 IEEE International Conference on Collaborative Advances in Software and COmputiNg (CASCON) (pp. 301-310). https://doi.org/10.1109/CASCON66301.2025.00056T. Zhang, L. Tahmooresnejad, and N. Ezzati-Jivan, "Energy Consumption Analysis of Large Language Models Across CPU and GPU Using Diverse Metric Types," in 2025 IEEE International Conference on Collaborative Advances in Software and COmputiNg (CASCON), pp. 301-310, 2025, doi: 10.1109/CASCON66301.2025.00056