diff options
| author | lijianfeng1993 <jianfeng.li@daocloud.io> | 2018-02-10 19:00:58 +0800 |
|---|---|---|
| committer | lijianfeng1993 <jianfeng.li@daocloud.io> | 2018-02-10 19:00:58 +0800 |
| commit | 3bc6b386f18146722cbbd22868e143ba831fcee1 (patch) | |
| tree | 85e1b01f1d69718834c108c8b199960e279e1966 | |
| parent | 36602d727348149e5cc3873f2bef0d582ad80ce4 (diff) | |
fix a duplication problem in gpu-support.md
| -rw-r--r-- | contributors/design-proposals/resource-management/gpu-support.md | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/contributors/design-proposals/resource-management/gpu-support.md b/contributors/design-proposals/resource-management/gpu-support.md index ad4383ac..2e74f4c3 100644 --- a/contributors/design-proposals/resource-management/gpu-support.md +++ b/contributors/design-proposals/resource-management/gpu-support.md @@ -75,7 +75,7 @@ Before any scheduling can occur, we need to know what's available out there. In v0, we'll hardcode capacity detected by the kubelet based on a flag, `--experimental-nvidia-gpu`. This will result in the user-defined resource `alpha.kubernetes.io/nvidia-gpu` to be reported for `NodeCapacity` and -`NodeAllocatable`, as well as as a node label. +`NodeAllocatable`, as well as a node label. ### Scheduling |