dyweb / papers-notebook

:page_facing_up: :cn: :page_with_curl: 论文阅读笔记(分布式系统、虚拟化、机器学习)Papers Notebook (Distributed System, Virtualization, Machine Learning)
https://github.com/dyweb/papers-notebook/issues?utf8=%E2%9C%93&q=is%3Aissue+is%3Aopen+-label%3ATODO-%E6%9C%AA%E8%AF%BB
Apache License 2.0
2.12k stars 244 forks source link

GaiaGPU: Sharing GPUs in Container Clouds #179

Open gaocegege opened 4 years ago

gaocegege commented 4 years ago

https://ieeexplore.ieee.org/abstract/document/8672318

腾讯 Gaia 的 GPU 共享

Ref #178

gaocegege commented 4 years ago

Kubernetes 部分非常容易理解

GPU 部分,采取了劫持 CUDA 调用的方式,进行了算力和内存的“隔离”,但具体怎么做的,写的并不清楚。这篇文章最核心的部分就是 vGPU Library 的实现这一块。

Screenshot from 2019-08-29 17-54-19

gaocegege commented 4 years ago

开源了 https://github.com/tkestack/vcuda-controller

IntelliJJ824 commented 2 years ago

我想问一下腾讯 Gaia的代码开源了吗?

gaocegege commented 2 years ago

开源了 https://github.com/tkestack/vcuda-controller

这里

cheimu commented 1 year ago

哈哈 最近正好想学习这个方面但找不到资源 关键时刻还得靠gaocegege 👯‍♂️

Zijie-Tian commented 1 year ago

nice

miyu233 commented 1 year ago

感谢