Skip to content

Conversation

@rpgreen
Copy link
Contributor

@rpgreen rpgreen commented Jan 13, 2026

Copy link
Contributor

@dantasse dantasse left a comment

Choose a reason for hiding this comment

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

Thanks Ryan!

Approving because it'd be fine if you just merged, some comments just to clarify things further.


- An existing Kubernetes cluster
- An existing node pool(s) for Geneva workloads. By default, Geneva uses node selector
`{"geneva.lancedb.com/ray-head": "true"}` for Ray head nodes, and
Copy link
Contributor

Choose a reason for hiding this comment

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

I think in our GCP cluster we've got {"geneva.lancedb.com/ray-head": ""} instead of "true". (same for the worker nodes) Is that just a quirk of our setup there?

Copy link
Contributor Author

Choose a reason for hiding this comment

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

That was fixed recently, both AWS and GCP follow this convention now


## Install

1. Authenticate with Kubernetes cluster
Copy link
Contributor

Choose a reason for hiding this comment

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

Sorry, this is super basic, but it's not obvious to me what this means in general. I think it means aws sso login or similar. If so, unless it's incredibly basic, want to add an example? (e.g.

1. Authenticate with Kubernetes cluster

(through your cloud service provider, like `aws sso login`, `gcp login`, or `azure login`)

Copy link
Contributor Author

Choose a reason for hiding this comment

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

fair point, will update

```
4. Install NVIDIA device plugin (if using GPU nodes)

For GPU support, the NVIDIA device plugin must be installed in your EKS cluster:
Copy link
Contributor

Choose a reason for hiding this comment

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

Is this the same on GCP and Azure?

Copy link
Contributor Author

Choose a reason for hiding this comment

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

Yeah

## Basic Kubernetes Setup

<Tip>
Kubernetes resources can be deployed automatically via [Helm](/geneva/deployment/helm/) or manually
Copy link
Contributor

Choose a reason for hiding this comment

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

Do we prefer the Helm installation? If so, we should say that, and probably have Helm before Manual in the sidebar. But if we don't care either way, then ignore this comment.

Copy link
Contributor Author

Choose a reason for hiding this comment

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

Yes we do. I kept this order because the helm install does depend on some cloud resources from this section.

Co-authored-by: Dan Tasse <105866+dantasse@users.noreply.github.com>
@rpgreen rpgreen merged commit bd46e7c into main Jan 13, 2026
2 checks passed
@rpgreen rpgreen deleted the rpgreen/geneva-helm-docs branch January 13, 2026 21:51
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.

2 participants