Additional kubernetes GPU docs #2078
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
webern
approved these changes
Apr 18, 2022
README.md
Outdated
| @@ -740,6 +741,11 @@ There are a few important caveats about the provided kdump support: | |||
| * The system kernel will reserve 256MB for the crash kernel, only when the host has at least 2GB of memory; the reserved space won't be available for processes running in the host | |||
| * The crash kernel will only be loaded when the `crashkernel` parameter is present in the kernel's cmdline and if there is memory reserved for it | |||
|
|
|||
| ### NVIDIA GPUs Support | |||
| Bottlerocket's `nvidia` kubernetes variants include the required packages and configurations to leverage NVIDIA GPUs. | |||
| The official AMIs for these variants can be used with the following EC2 instance types: `p2`, `p3`, `p4`, `g4dn`, `g5` and `g5g`. | |||
There was a problem hiding this comment.
Suggestion: Guard against having an incorrect list in the future.
Suggested change
| The official AMIs for these variants can be used with the following EC2 instance types: `p2`, `p3`, `p4`, `g4dn`, `g5` and `g5g`. | |
| The official AMIs for these variants can be used with EC2 GPU-equipped instance types such as: `p2`, `p3`, `p4`, `g4dn`, `g5` and `g5g`. |
jpculp
reviewed
Apr 18, 2022
QUICKSTART-EKS.md
Outdated
| @@ -383,3 +383,20 @@ You can install them in your cluster by following the `helm install` instruction | |||
| The [GPU Operator](https://docs.nvidia.com/datacenter/cloud-native/gpu-operator/getting-started.html#install-nvidia-gpu-operator) can also be used to install these tools. | |||
| However, it is cumbersome to select the right subset of features to avoid conflicts with the software included in the variant. | |||
| Therefore we recommend installing the tools individually if they are required. | |||
|
|
|||
| In hosts with multiple GPUs (i.e. EC2 `g4dn` instances) you can assign a GPU per container by specifying the resource in the containers' spec as described in the [official kubernetes documentation](https://kubernetes.io/docs/tasks/manage-gpus/scheduling-gpus/): | |||
There was a problem hiding this comment.
Suggested change
| In hosts with multiple GPUs (i.e. EC2 `g4dn` instances) you can assign a GPU per container by specifying the resource in the containers' spec as described in the [official kubernetes documentation](https://kubernetes.io/docs/tasks/manage-gpus/scheduling-gpus/): | |
| In hosts with multiple GPUs (ex. EC2 `g4dn` instances) you can assign a GPU per container by specifying the resource in the containers' spec as described in the [official kubernetes documentation](https://kubernetes.io/docs/tasks/manage-gpus/scheduling-gpus/): |
e.g. would also work.
arnaldo2792
force-pushed
the
k8s-gpu-docs
branch
from
5be3daf
to
a8c216f
Compare
|
Forced push addresses the comments above |
bcressey
reviewed
Apr 21, 2022
README.md
Outdated
| @@ -740,6 +741,11 @@ There are a few important caveats about the provided kdump support: | |||
| * The system kernel will reserve 256MB for the crash kernel, only when the host has at least 2GB of memory; the reserved space won't be available for processes running in the host | |||
| * The crash kernel will only be loaded when the `crashkernel` parameter is present in the kernel's cmdline and if there is memory reserved for it | |||
|
|
|||
| ### NVIDIA GPUs Support | |||
| Bottlerocket's `nvidia` kubernetes variants include the required packages and configurations to leverage NVIDIA GPUs. | |||
There was a problem hiding this comment.
nit: maybe omit Kubernetes since we'll have ECS soon:
Suggested change
| Bottlerocket's `nvidia` kubernetes variants include the required packages and configurations to leverage NVIDIA GPUs. | |
| Bottlerocket's `nvidia` variants include the required packages and configurations to leverage NVIDIA GPUs. |
bcressey
reviewed
Apr 21, 2022
README.md
Outdated
| ### NVIDIA GPUs Support | ||
| Bottlerocket's `nvidia` kubernetes variants include the required packages and configurations to leverage NVIDIA GPUs. | ||
| The official AMIs for these variants can be used with EC2 GPU-equipped instance types such as: `p2`, `p3`, `p4`, `g4dn`, `g5` and `g5g`. | ||
| Please refer to the [Amazon EKS quickstart](QUICKSTART-EKS.md#aws-k8s--nvidia-variants) for further details about these variants. |
There was a problem hiding this comment.
Elsewhere in this doc, we refer to this doc as QUICKSTART-EKS so I'd like to continue using that name for consistency.
This adds a new section for NVIDIA GPUs and lists what EC2 instance types are supported by the official Bottlerocket `nvidia` k8s AMIs. Signed-off-by: Arnaldo Garcia Rincon <agarrcia@amazon.com>
arnaldo2792
force-pushed
the
k8s-gpu-docs
branch
from
a8c216f
to
506e55a
Compare
|
Forced push addresses comments above |
bcressey
approved these changes
Apr 22, 2022
README.md
Outdated
| ### NVIDIA GPUs Support | ||
| Bottlerocket's `nvidia` variants include the required packages and configurations to leverage NVIDIA GPUs. | ||
| The official AMIs for these variants can be used with EC2 GPU-equipped instance types such as: `p2`, `p3`, `p4`, `g4dn`, `g5` and `g5g`. | ||
| Please see [QUICKSTART-EKS](QUICKSTART-EKS.md#aws-k8s--nvidia-variants) for further details about kubernetes variants. |
There was a problem hiding this comment.
Suggested change
| Please see [QUICKSTART-EKS](QUICKSTART-EKS.md#aws-k8s--nvidia-variants) for further details about kubernetes variants. | |
| Please see [QUICKSTART-EKS](QUICKSTART-EKS.md#aws-k8s--nvidia-variants) for further details about Kubernetes variants. |
Now the documentation explicitly says that it is possible to use a GPU per orchestrated container, and references the official kubernetes documentation to schedule NVIDIA GPUs. Signed-off-by: Arnaldo Garcia Rincon <agarrcia@amazon.com>
arnaldo2792
force-pushed
the
k8s-gpu-docs
branch
from
506e55a
to
d398824
Compare
|
(Forced push fixes comment above) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Issue number:
N / A
Description of changes:
Testing done:
Terms of contribution:
By submitting this pull request, I agree that this contribution is dual-licensed under the terms of both the Apache License, version 2.0, and the MIT license.