低碳AI:大模型的绿色训练与推理优化方法研究

doi:10.12267/j.issn.2096-5931.2025.06.008

信息通信技术与政策 ›› 2025, Vol. 51 ›› Issue (6): 44-51.doi: 10.12267/j.issn.2096-5931.2025.06.008

低碳AI:大模型的绿色训练与推理优化方法研究

Low-carbon AI: research on green training and inference optimization methods for large model

葛坚¹, 牛晓燕², 毕然², 黄雍涛²

1.中国信息通信研究院技术与标准研究所,北京 100191
2.中国信息通信研究院安全研究所,北京 100191

收稿日期:2025-04-22 出版日期:2025-06-25 发布日期:2025-07-04
通讯作者: 毕然中国信息通信研究院安全研究所高级工程师,主要从事AI赋能通信大数据平台等研究工作
作者简介:
葛坚中国信息通信研究院技术与标准研究所高级工程师,主要从事智算网络、通信核心网相关研究工作
牛晓燕中国信息通信研究院安全研究所工程师,主要从事AI赋能通信大数据平台等研究工作
黄雍涛中国信息通信研究院安全研究所工程师,主要从事AI赋能通信大数据平台等研究工作

GE Jian¹, NIU Xiaoyan², BI Ran², HUANG Yongtao²

1. Institute of Technology and Standards, China Academy of Information and Communications Technology, Beijing 100191, China
2. Institute of Security Research, China Academy of Information and Communications Technology, Beijing 100191, China

Received:2025-04-22 Online:2025-06-25 Published:2025-07-04

摘要/Abstract

摘要：

近年来,人工智能(Artificial Intelligence,AI)模型的规模不断扩大,训练与推理过程的能耗问题日益突出,推动了低碳AI研究的兴起。系统梳理了大模型绿色计算的关键技术,重点分析了低碳训练与推理优化方法。在训练阶段,现有研究主要从模型架构优化、计算精度调整、资源调度策略等方面降低能耗,包括神经网络架构搜索、混合精度计算及分布式训练等方法。在推理阶段,优化策略则侧重于模型剪枝、量化技术、边缘计算、缓存复用等技术,以减少计算成本与碳排放。此外,归纳了低碳AI面临的挑战,如计算硬件能效瓶颈、碳足迹量化方法的不确定性以及数据中心绿色能源利用的限制,并探讨了未来的发展趋势。

关键词: 大语言模型, 人工智能, 低碳, 绿色计算

Abstract:

In recent years, the scale of Artificial Intelligence (AI) models has been expanding continuously, leading to increasing energy consumption in training and inference processes. This has driven the rise of low-carbon AI research. This paper systematically reviews the key technologies of green computing for large models, with a focus on low-carbon training and inference optimization methods. In the training phase, existing studies aim to reduce energy consumption through model architecture optimization, computational precision adjustment, and resource scheduling strategies. These approaches include neural architecture search, mixed-precision computing, and distributed training. In the inference phase, optimization strategies primarily focus on techniques such as model pruning, quantization, edge computing, and cache reuse to reduce computational costs and carbon emissions. Additionally, this paper summarizes the challenges faced by low-carbon AI, including hardware energy efficiency bottlenecks, uncertainties in carbon footprint quantification methods, and limitations in the use of green energy in data centers. Furthermore, this paper explores the future development trends.

Key words: large language model, AI, low carbon, green computing

中图分类号:

TN915.5

葛坚, 牛晓燕, 毕然, 黄雍涛. 低碳AI:大模型的绿色训练与推理优化方法研究[J]. 信息通信技术与政策, 2025, 51(6): 44-51.

GE Jian, NIU Xiaoyan, BI Ran, HUANG Yongtao. Low-carbon AI: research on green training and inference optimization methods for large model[J]. Information and Communications Technology and Policy, 2025, 51(6): 44-51.

导出引用管理器 EndNote|Ris|BibTeX

链接本文:

http://ictp.caict.ac.cn/CN/10.12267/j.issn.2096-5931.2025.06.008

http://ictp.caict.ac.cn/CN/Y2025/V51/I6/44

图/表 5

参考文献 29

[1]	BROWN T, MANN B, RYDER N, et al. Language models are few-shot learners[C]//Proceedings of the 34th International Conference on Neural Information Processing Systems(NIPS’20). Red Hook: Curran Associates Inc., 2020, 33: 1877-1901.
[2]	中央网络安全和信息化委员会. “十四五”国家信息化规划[EB/OL], 2021[2025-04-22]. https://www.gov.cn/xinwen/2021-12/28/5664873/files/1760823a103e4d75ac681564fe481af4.pdf.
[3]	国家能源局. 国家能源局关于加快推进能源数字化智能化发展的若干意见[EB/OL]. (2023-03-28)[2025-04-22]. https://www.gov.cn/zhengce/zhengceku/2023-04/02/content_5749758.htm.
[4]	梁芳, 佟恬, 马贺荣, 等. 东数西算下算力网络发展分析[J]. 信息通信技术与政策, 2022, 48(11):79-83. doi: 10.12267/j.issn.2096-5931.2022.11.010
[5]	Stanford University Human-Centered AI. Artificial intelligence index report 2023[R], 2023.
[6]	ZHOU Y, LIN X J, ZHANG X, et al. On the opportunities of green computing: a survey[J]. arXiv Preprint, arXiv:2311.00447, 2023.
[7]	ZOIE R C, MIHAELA R D, ALEXANDRU S. An analysis of the power usage effectiveness metric in data centers[C]//2017 5th International symposium on electrical and electronics engineering (ISEEE). Galati: IEEE, 2017:1-6.
[8]	PRICE D C, CLARK M A, BARSDELL B R, et al. Optimizing performance-per-watt on GPUs in high performance computing: temperature, frequency and voltage effects[J]. Computer Science-Research and Development, 2016, 31(4): 185-193.
[9]	陈巍. 大模型技术与产业分析[EB/OL]. 2024[2025-04-22]. https://www.zhihu.com/column/c_1607003152053755904.
[10]	PATEL J M. Introduction to common crawl datasets[M]// Getting structured data from the internet:running web crawlers/scrapers on a big data production scale. Berkeley: Apress, 2020: 277-324.
[11]	SUN Z, YU H, SONG X, et al. Mobilebert: a compact task-agnostic bert for resource-limited devices[J]. arXiv Preprint, arXiv:2004. 02984, 2020.
[12]	JIAO X, YIN Y, SHANG L, et al. Tinybert: distilling bert for natural language understanding[J]. arXiv Preprint, arXiv:1909.10351, 2019.
[13]	FEDUS W, ZOPH B, SHAZEER N. Switch transformers: scaling to trillion parameter models with simple and efficient sparsity[J]. Journal of Machine Learning Research, 2022, 23(120):1-39.
[14]	LIU H, LI Z, HALL D, et al. Sophia: a scalable stochastic second-order optimizer for language model pre-training[J]. arXiv Preprint, arXiv:2305.14342, 2023.
[15]	LIANG H, ZHANG S, SUN J, et al. Darts+: improved differentiable architecture search with early stopping[J]. arXiv Preprint, arXiv:1909.06035, 2019.
[16]	HARLAP A, NARAYANAN D, PHANISHAYEE A, et al. Pipedream: fast and efficient pipeline parallel dnn training[J]. arXiv Preprint, arXiv:1806.03377, 2018.
[17]	LIN Y, HAN S, MAO H, et al. Deep gradient compression: reducing the communication bandwidth for distributed training[J]. arXiv Preprint,arXiv:1712.01887, 2017.
[18]	TANG H, YU C, LIAN X, et al. Doublesqueeze: parallel stochastic gradient descent with double-pass error-compensated compression[C]// International Conference on Machine Learning. PMLR, 2019: 6155-6165.
[19]	LIN Z, COURBARIAUX M, MEMISEVIC R, et al. Neural networks with few multiplications[J]. arXiv Preprint, arXiv:1510.03009, 2015.
[20]	MICIKEVICIUS P, NARANG S, ALBEN J, et al. Mixed precision training[J]. arXiv Preprint, arXiv:1710.03740, 2017.
[21]	HAN S, MAO H, DALLY W J. Deep compression: compressing deep neural networks with pruning, trained quantization and huffman coding[J]. arXiv Preprint, arXiv:1510.00149, 2015.
[22]	WANG X, TANG Z, GUO J, et al. Empowering edge intelligence: a comprehensive survey on on-device ai models[J]. ACM Computing Surveys, 2025, 57(9):39.
[23]	TAN M, LE Q. Efficientnet: rethinking model scaling for convolutional neural networks[C]//International conference on machine learning. PMLR,2019: 6105-6114.
[24]	GHOSH S K, RAHA A, RAGHUNATHAN V. Energy-efficient approximate edge inference systems[J]. ACM Transactions on Embedded Computing Systems, 2023, 22(4): 1-50.
[25]	任杰, 高岭, 于佳龙, 等. 面向边缘设备的高能效深度学习任务调度策略[J]. 计算机学报, 2020, 43(3): 440-452.
[26]	ZHAI Y, JIANG C, WANG L, et al. Bytetransformer:a high-performance transformer boosted for variable-length inputs[C]//2023 IEEE International Parallel and Distributed Processing Symposium (IPDPS). Atlanta: IEEE, 2023: 344-355.
[27]	WU B, LIU S, ZHONG Y, et al. Loongserve:efficiently serving long-context large language models with elastic sequence parallelism[C]//Proceedings of the ACM SIGOPS 30th Symposium on Operating Systems Principles(SOSP’24). New York: Association for Computing Machinery, 2024: 640-654.
[28]	SONG Y, MI Z, XIE H, et al. Powerinfer: fast large language model serving with a consumer-grade GPU[C]//Proceedings of the ACM SIGOPS 30th Symposium on Operating Systems Principles. New York: Association for Computing Machinery, 2024: 590-606.
[29]	XU M, ZHU M, LIU Y, et al. Deepcache: principled cache for mobile deep vision[C]//Proceedings of the 24th annual international conference on mobile computing and networking. New York: Association for Computing Machinery, 2018:129-144.

GPU型号	浮点运算次数 /TFLOPs	额定功率 /W	计算效率/ GFLOPs/W
NVIDIA H100	51.0	350	145.71
NVIDIA A100	19.5	400	48.75
NVIDIA V100	7.0	300	23.30

GPU型号	浮点运算次数 /TFLOPs	额定功率 /W	计算效率/ GFLOPs/W
NVIDIA H100	51.0	350	145.71
NVIDIA A100	19.5	400	48.75
NVIDIA V100	7.0	300	23.30

大模型	输入 tokens/$	输入tokens (缓存命中)/$	输出 tokens/$
DeepSeek-Chat	0.27	0.07	1.10
GPT-4o	2.50	1.25	10.00
GPT-4.5-preview	75.00	37.50	150.00

大模型	输入 tokens/$	输入tokens (缓存命中)/$	输出 tokens/$
DeepSeek-Chat	0.27	0.07	1.10
GPT-4o	2.50	1.25	10.00
GPT-4.5-preview	75.00	37.50	150.00

低碳AI:大模型的绿色训练与推理优化方法研究

Low-carbon AI: research on green training and inference optimization methods for large model

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 5

参考文献 29

相关文章 15

编辑推荐

Metrics

本文评价

[1]	岳锦疆, 宁庭勇. 人工智能在固态储氢材料中的综合应用[J]. 信息通信技术与政策, 2025, 51(6): 21-26.
[2]	杨光, 尹会会, 凌大兵, 韩笑. 人工智能赋能石化行业转型升级研究[J]. 信息通信技术与政策, 2025, 51(6): 33-37.
[3]	汪振涛. 人工智能驱动能源系统变革的挑战与应对策略[J]. 信息通信技术与政策, 2025, 51(6): 38-43.
[4]	王宇航, 史翔, 申一彤. 能源行业AI大模型工程化实践研究[J]. 信息通信技术与政策, 2025, 51(6): 52-59.
[5]	王琦瑶. 人工智能技术在电池研发领域的专利情报分析[J]. 信息通信技术与政策, 2025, 51(6): 60-65.
[6]	安成, 王静宜, 逯峰. 基于新一代信息技术提升煤矿物联感知和控制交互能力研究^*[J]. 信息通信技术与政策, 2025, 51(6): 66-72.
[7]	黄树巍, 金枫, 陈继巍, 赵崇岩. 人工智能赋能矿山生产智能创新发展[J]. 信息通信技术与政策, 2025, 51(6): 8-14.
[8]	赵远, 张帅, 于青, 张奕, 徐彬, 杨旭. NFV云计算绿色发展趋势及关键技术研究[J]. 信息通信技术与政策, 2025, 51(5): 12-21.
[9]	李贺, 冯天昊. 绿色化对我国工业增长的影响与转型路径分析[J]. 信息通信技术与政策, 2025, 51(5): 22-27.
[10]	祁航, 杨吉双, 刘家祺. 我国工业领域低碳技术革新路径[J]. 信息通信技术与政策, 2025, 51(5): 43-50.
[11]	杨吉双, 王华丽, 杨晨. 人工智能时代智能算力带来的能源挑战与建议[J]. 信息通信技术与政策, 2025, 51(5): 58-62.
[12]	李腾达, 邬知衡. 数字技术赋能绿色低碳发展:理论、实践与挑战[J]. 信息通信技术与政策, 2025, 51(5): 63-67.
[13]	卢春阳, 祁航, 郝昫, 刘妍. ICT行业低碳发展成效与策略建议[J]. 信息通信技术与政策, 2025, 51(5): 75-79.
[14]	杨璐, 刘家祺, 卢春阳. 信息通信行业绿色供应链管理研究[J]. 信息通信技术与政策, 2025, 51(5): 80-84.
[15]	刘高峰, 姜约瀚, 张育雄. 人工智能赋能数字孪生城市高水平发展和高质量建设[J]. 信息通信技术与政策, 2025, 51(3): 74-82.