NVIDIA open-sources cuFile API to advance direct GPU-to-storage access
Original title:As AI Increases Demands on Memory, Storage Steps Up
AI summary
At the FMS conference, NVIDIA announced the open-sourcing of the cuFile API and the underlying storage software stack, enabling GPUs to read and write storage directly with access latency reduced to microseconds. Its Vera CPU delivers up to 3.21 times the throughput of x86 CPUs in two-stage compression and encryption pipelines, and the company launched the Storage-Next initiative with more than 40 vendors.
Key Takeaways
- At the FMS conference,NVIDIA announced the open-sourcing of the cuFile API and the underlying…
- Its Vera CPU delivers up to 3.
- 21 times the throughput of x86 CPUs in two-stage compression and encryption pipelines,and the…
Quick Facts
- Published
- Updated
- Source
- NVIDIA Blog(RSS)
- Category
- Products & Apps
- AI score
- 62
cuFile 开源让 GPU 直存访问成为可跨平台协作的开放标准,这会改变 AI 基础设施团队评估存储架构的方式,不再只绑定单一厂商的实现。
Related entities
Source & copyright
This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.
How to Cite
When citing this article, keep the title, date, and permalink, and credit Orange News as the source.
Orange News: NVIDIA open-sources cuFile API to advance direct GPU-to-storage access. August 4, 2026. https://new.nat6.net/en/news/cmsesrlcn15j7ro2evd57my2c/