arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2407.04014cs.DC
收藏

离线能量最优的LLM服务:面向异构系统上LLM推理的基于工作负载的能量模型

Offline Energy-Optimal LLM Serving: Workload-Based Energy Models for LLM Inference on Heterogeneous Systems

Grant Wilkins, Srinivasan Keshav, Richard Mortier

AI总结:

本文针对异构GPU-CPU系统上的LLM推理能耗问题,通过特性表征建立高精度能量与运行时间模型,并提出离线能量最优调度框架,验证了其优于现有实践。

摘要:

大语言模型(LLMs)的快速采用推动了自然语言处理和文本生成领域的显著进步。然而,LLM模型推理所消耗的能量仍是可持续AI部署面临的主要挑战。为解决这一问题,我们对异构GPU-CPU系统上LLM推理任务中与工作负载相关的能量消耗和运行时间进行建模。通过对多个最先进的LLM开展广泛的特性表征研究,并分析它们在不同规模的输入提示和输出文本下的能量与运行时间行为,我们为每个LLM建立了准确的(R^2>0.96)能量和运行时间模型。我们利用这些模型探索了一个离线的、能量最优的LLM工作负载调度框架。通过案例研究,我们展示了与现有最佳实践相比,兼顾能量和准确率感知的调度所具有的优势。

英文摘要:

The rapid adoption of large language models (LLMs) has led to significant advances in natural language processing and text generation. However, the energy consumed through LLM model inference remains a major challenge for sustainable AI deployment. To address this problem, we model the workload-dependent energy consumption and runtime of LLM inference tasks on heterogeneous GPU-CPU systems. By conducting an extensive characterization study of several state-of-the-art LLMs and analyzing their energy and runtime behavior across different magnitudes of input prompts and output text, we develop accurate (R^2>0.96) energy and runtime models for each LLM. We employ these models to explore an offline, energy-optimal LLM workload scheduling framework. Through a case study, we demonstrate the advantages of energy and accuracy aware scheduling compared to existing best practices.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑