英文:
How set maximum concurrent requests per instance in Kubernetes to 1?
问题
我想为每个用户分配每个虚拟机。
在 Cloud Run 中,只需将最大并发请求设置为 1,就可以正常工作,但 Cloud Run 不支持 GPU,对我来说不是一个选项。
在 Kubernetes 上如何实现?我正在在虚拟机上运行虚幻引擎,不应该共享。
英文:
I want to dedicate each vm to each user.
In cloud run it is simply working by set maximum concurrent requests to 1, however cloud run does not support gpu and it is not an option for me.
How is it possible on Kubernetes? I am running Unreal Engine on VMs and this should not be shared.
答案1
得分: 3
没有开箱即用的Kubernetes。但安装KNative,您将能够实现它(并且您将能够重用与Cloud Run相同的YAML服务定义,因为Cloud Run实现了KNative API)。
英文:
No Kubernetes out of the box. But install KNative and you will be able to achieve it (and you will be able to reuse the same YAML service definition as Cloud Run, because Cloud Run implements the KNative APIs)
通过集体智慧和协作来改善编程学习和解决问题的方式。致力于成为全球开发者共同参与的知识库,让每个人都能够通过互相帮助和分享经验来进步。
评论