Skip to content

Conversation

@almaslennikov
Copy link
Collaborator

@almaslennikov almaslennikov commented Jun 17, 2025

  • Add a new quick start section with (almost) ready-to-use deployment examples
  • Add keywords to use case definitions for ease of future automated parsing by LLMs
  • Store examples as separate files and build them to allow automated deployments

TBD:

  • Test the examples
  • Allow to change other parameters in deployment examples (e.g. pci addresses or interface names)

Copy link
Member

@rollandf rollandf left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Great work!

@almaslennikov almaslennikov marked this pull request as ready for review June 25, 2025 08:07
Copy link

@thuffmlx thuffmlx left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is great.
I would add at least 1-3 more 2ndary RDMA network definitions.
Most configurations have 4-8 NICS and consequently 4-8 parallel secondary RDMA networks. Would help clarify the list of networks in the annotations, and the corresponding list of resource requests. I've encountered customers that tried to dump all of the NICs in to the same secondary network as a pool.

@almaslennikov
Copy link
Collaborator Author

This is great. I would add at least 1-3 more 2ndary RDMA network definitions. Most configurations have 4-8 NICS and consequently 4-8 parallel secondary RDMA networks. Would help clarify the list of networks in the annotations, and the corresponding list of resource requests. I've encountered customers that tried to dump all of the NICs in to the same secondary network as a pool.

Hey @thuffmlx, please take a look at the new version. I've added more RDMA network definitions to IPoIB RDMA shared use case

Copy link
Member

@rollandf rollandf left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Great job!
Thanks

@github-actions
Copy link

}
nvIpam:
image: nvidia-k8s-ipam
repository: |nvidia-ipam-repository|
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

just a suggestions but you could use something more "standard"

like using envsubst instead of needing to generate sed cmd for each var in vars.rst.

that means

  1. have var.env file created from release.yaml
  2. generate vars.rsh from 1
  3. generate examples using vars from 1

example: in templates you will put something like: repository: $NVIDIA_IPAM_REPO

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Good idea.
let's fix this in a follow up

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Issue created:
#238

* Add a new quick start section with (almost) ready-to-use deployment examples
* Add keywords to use case definitions for ease of future automated parsing
* Store examples as separate files and build them to allow automated deployments
* Rename getting started with k8s to deployment guide and put it on the same level with quick start guide

Signed-off-by: Alexander Maslennikov <amaslennikov@nvidia.com>
@rollandf rollandf merged commit 4fc9627 into Mellanox:main Jul 29, 2025
2 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants