English(EN) Dissecting GPU Utilization for LLM Inference on Nvidia Hopper

论文剖析Nvidia Hopper上LLM推理的GPU利用率

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-14 04:00

一篇新论文探讨了Nvidia Hopper架构上大型语言模型（LLM）推理的GPU利用率的复杂性。研究强调，单一的利用率百分比可能具有误导性，掩盖了在解码操作期间的低效率，此时小批量请求仅部分填充计算片段。该研究提出使用源自Nsight Compute报告的八个不同、经过反向验证的视图，将利用率差距映射到诸如片段填充、占用率限制和内核选择等特定机制。 AI

影响为优化高端GPU上的LLM推理性能提供了更深入的见解，可能导致更高效的部署。

排序理由该集群包含一篇详细介绍LLM推理性能技术研究的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.LG 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 27 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇详细介绍LLM推理性能技术研究的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, infra; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.LG TIER_1 English(EN) · Mohammad Siavashi, Gerald Q. Maguire Jr., Dejan Kostic, Marco Chiesa · 2026-09-14 04:00

解析Nvidia Hopper上LLM推理的GPU利用率

arXiv:2609.12923v1 Announce Type: cross Abstract: A single SM utilization percentage can make an LLM inference workload look compute-saturated while hiding how much useful work is being done. The problem is not that the counter is wrong, but that it collapses several different me…

本文如何被排名

报道来源 [1]

解析Nvidia Hopper上LLM推理的GPU利用率

相关实体

相关话题