JupyterLab Workload¶

This workload deploys a basic JupyterLab environment on top of any image with Python (pip) pre-installed. It is ideal for interactive development sessions and experimentation with other workloads.

Configuration Parameters¶

You can configure the following parameters in the values.yaml file or override them via the command line:

Parameter	Description	Default
`image`	Container image repository and tag	`rocm/pytorch:rocm6.4_ubuntu24.04_py3.12_pytorch_release_2.6.0`
`imagePullPolicy`	Image pull policy	`Always`
`gpus`	Number of GPUs to allocate (set to 0 for CPU-only mode)	`1`
`memory_per_gpu`	Memory allocated per GPU (in Gi)	`128`
`cpu_per_gpu`	CPU cores allocated per GPU	`4`
`storage.ephemeral`	Ephemeral storage configuration	`128Gi`, `mlstorage`, `ReadWriteOnce`
`deployment.ports.http`	HTTP port exposed by the service	`8080`
`entrypoint`	Custom entrypoint script	See `values.yaml` for details

For more details see the values.yaml file.

Deploying the Workload¶

To deploy the chart with the release name example, run the following command from the helm/ directory:

helm template example . | kubectl apply -f -

Note: If you set the gpus value greater than 0, ensure you specify a GPU-capable image to utilize the allocated resources properly.

Accessing the Workload Locally¶

To access JupyterLab locally, forward the service port to your machine:

kubectl port-forward services/dev-workspace-jupyterlab-example 8080:80

Then, open your browser and navigate to http://localhost:8080.

Accessing the Workload via URL¶

To access the workload through a URL, you can enable either an Ingress or HTTPRoute in the values.yaml file. The following parameters are available:

Parameter	Description	Default
`ingress.enabled`	Enable Ingress resource	`false`
`httproute.enabled`	Enable HTTPRoute resource	`false`

See the corresponding template files in the templates/ directory. For more details on configuring Ingress or HTTPRoute, refer to the Ingress documentation and HTTPRoute documentation, or documentation of the particular gateway implementation you may use, like KGateway. Check with your cluster administrator for the correct configuration for your environment.