Add GPU edge node manege docs #654
Conversation
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: The full list of commands accepted by this bot can be found here.
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
kubeedge-bot
added
the
size/XL
wbc6080
force-pushed
the
add-gpu-docs
branch
from
399a860 to
d7fb3be
Compare
|
/lgtm |
kubeedge-bot
added
lgtm
|
New changes are detected. LGTM label has been removed. |
wbc6080
force-pushed
the
add-gpu-docs
branch
3 times, most recently
from
47188ae to
0fa121e
Compare
Signed-off-by: wbc6080 <[email protected]> Co-authored-by: ming.tang <[email protected]>
|
PTAL @fisherxu @Shelley-BaoYue |
Signed-off-by: wbc6080 <[email protected]>
|
|
||
| ## Abstract | ||
|
|
||
| With the development of edge AI, the demand for deploying GPU applications on edge nodes is gradually increasing. Currently, KubeEdge can manage GPU nodes through some configurations, |
There was a problem hiding this comment.
| With the development of edge AI, the demand for deploying GPU applications on edge nodes is gradually increasing. Currently, KubeEdge can manage GPU nodes through some configurations, | |
| With the development of edge AI, it is likely that the applications at edge demand and rely on the GPU acceleration. Currently, KubeEdge can manage GPU nodes through some configurations, |
| ## Abstract | ||
|
|
||
| With the development of edge AI, the demand for deploying GPU applications on edge nodes is gradually increasing. Currently, KubeEdge can manage GPU nodes through some configurations, | ||
| and allocate GPU resources to user edge applications through the k8s device-plugin component. If you need to use this feature, please refer to the steps below. |
There was a problem hiding this comment.
| and allocate GPU resources to user edge applications through the k8s device-plugin component. If you need to use this feature, please refer to the steps below. | |
| and allocate GPU resources to user edge applications through the [k8s device-plugin](https://kubernetes.io/docs/concepts/extend-kubernetes/compute-storage-net/device-plugins/) component. If you need to use this feature, please refer to the steps below. |
|
|
||
| 1. Install GPU driver | ||
|
|
||
| First you need to determine whether the edge node machine has GPU. You can use the `lspci | grep NVIDIA` command to check. Download the appropriate GPU driver according to the specific GPU model and complete the installation. |
There was a problem hiding this comment.
so it turned out that this is only for Nvidia GPUs? if that is so, probably we could change the title into How to enable Nvidia GPUs on KubeEdge or something else?
|
|
||
| ## Getting Started | ||
|
|
||
| ### GPU running environment construction |
There was a problem hiding this comment.
Can we instead just point some URLs such as https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/install-guide.html to make sure Nvidia GPU is enabled and configured for container runtime?
This is obviously not the procedure KubeEdge provides, to avoid having the maintenance cost to catch up with Nvidia specific procedure, it would be easier for the doc maintainers?
|
|
||
| Hosting edge GPU nodes mainly includes the following steps: | ||
|
|
||
| 1. Manage the node to the cluster |
There was a problem hiding this comment.
same here can we just point the reference to https://kubeedge.io/docs/setup/install-with-keadm/ to avoid the redundancy and possible maintenance problem?
Which issue(s) this PR fixes:
Fixes #
docs update
Added documentation for managing edge GPU nodes and introduce how to use GPU resources in edge applications.