Deploying on Kubernetes#

The Helm chart simplifies NIM deployment on Kubernetes. The chart supports deployment with a variety of possible cluster, GPU and storage confurations. The chart downloads the model and starts up the service to begin running.

NIMs are intended to be run on a system with NVIDIA GPUs, with the type and number of GPUs depending on the model. To use helm, you must have a Kubernetes cluster with appropriate GPU nodes and the GPU Operator installed.

Benefits of Helm Chart Deployment#

Using a helm chart:

  • Enables using Kubernetes Nodes and horizontally scaling the service

  • Encapsulates the complexity of running Docker commands directly

  • Enables monitoring metrics from the NIM

Setting Up the Environment#

If you haven’t set up your NGC API key and do not know exactly which NIM you want to download and deploy, see the information in the User Guide.

This helm chart requires that you have a secret with your NGC API key configured for downloading private images, and one with your NGC API key (below named ngc-api). These will likely have the same key in it, but they will have different formats (dockerconfig.json vs opaque). See Creating Secrets below.

These instructions will assume that you have your NGC_API_KEY exported in the environment.

export NGC_API_KEY="<YOUR NGC API KEY>"

Fetching the Helm Chart#

You can fetch the helm chart from NGC by executing the following command:

helm fetch https://helm.ngc.nvidia.com/nim/nvidia/charts/nvidia-nim-llama-32-nv-embedqa-1b-v2-1.3.0.tgz --username='$oauthtoken' --password=$NGC_API_KEY

You can use OpenTelemetry for monitoring your container. See OpenTelemetry parameters for details.

Namespace#

You can choose to deploy to whichever namespace is appropriate, but for documentation purposes we will deploy to a namespace named embedding-nim.

kubectl create namespace embedding-nim

Creating Secrets#

Use the following procedure to create the expected secrets for this helm chart.

  1. Run the following code to set your NGC registry password.

    DOCKER_CONFIG='{"auths":{"nvcr.io":{"username":"$oauthtoken", "password":"'${NGC_API_KEY}'" }}}'
echo -n $DOCKER_CONFIG | base64 -w0
NGC_REGISTRY_PASSWORD=$(echo -n $DOCKER_CONFIG | base64 -w0 )

  2. Run the following code to create the secrets.

    kubectl apply -n embedding-nim -f - <<EOF
apiVersion: v1
kind: Secret
metadata:
  name: ngc-secret
type: kubernetes.io/dockerconfigjson
data:
  .dockerconfigjson: ${NGC_REGISTRY_PASSWORD}
EOF

kubectl create -n embedding-nim secret generic ngc-api --from-literal=NGC_API_KEY=${NGC_API_KEY}

Configuration Considerations#

The following deployment commands will by default create a single deployment with one replica using the llama-3.2-nv-embedqa-1b-v2 model. The following options can be used to make modifications to the behavior. See Parameters for a description of the Helm parameters.

  • image.repository – The container (llama-3.2-nv-embedqa-1b-v2) to deploy

  • image.tag – The version of that container (1.3.0)

  • Storage options, based on the environment and cluster in use

  • resources – Use this option when a model requires more than the default of one GPU. See below for support matrix and resource requirements.

  • env – Which is an array of environment variables presented to the container, if advanced configuration is needed

Storage#

This NIM uses persistent storage for storing downloaded models. These instructions require that you have a local-nfs storage class provisioner installed in your cluster.

helm repo add nfs-ganesha-server-and-external-provisioner https://kubernetes-sigs.github.io/nfs-ganesha-server-and-external-provisioner/
helm install nfs-server nfs-ganesha-server-and-external-provisioner/nfs-server-provisioner --set storageClass.name=local-nfs

Advanced Storage Configuration#

Storage is a particular concern when setting up NIMs. Models can be quite large, and you can fill a disk downloading things to emptyDirs or other locations around your pod image. We recommend that you mount persistent storage of some kind on your pod.

This chart supports two general categories:

  1. Persistent Volume Claims (enabled with persistence.enabled)

  2. hostPath (enabled with persistences.hostPath)

By default, the chart uses the standard storage class and creates a PersistentVolume and a PersistentVolumeClaim.

If you do not have a Storage Class Provisioner that creates PersistentVolumes automatically, set the value persistence.createPV=true. This is also necessary when you use persistence.hostPath on minikube.

If you have an existing PersistentVolumeClaim where you’d like the models to be stored at, pass that value in at persistence.exsitingClaimName.

See the Helm options in Parameters.

Deploying#

Use the following bash command to create a basic deployment.

helm upgrade --install \
  --namespace embedding-nim \
  nemo-embedder \
  --set persistence.class="local-nfs" \
  nvidia-nim-llama-32-nv-embedqa-1b-v2-1.3.0.tgz

You can also change the version of the model in use by adding the following after --namespace

--set image.tag=1.3.0 \

After deploying check the pods to ensure that it is running, initial image pull and model download can take upwards of 15 minutes.

kubectl get pods -n embedding-nim

The pod should eventually end up in the running state.

NAME              READY   STATUS    RESTARTS   AGE
nvidia-nim-llama-32-nv-embedqa-1b-v2-0   1/1     Running   0          8m44s

Check events for failures:

kubectl get events -n embedding-nim

Deploying Snowflake Arctic Embedding#

Run the helm command with the following parameters, update your version in image.tag:

helm fetch https://helm.ngc.nvidia.com/nim/nvidia/charts/text-embedding-nim-1.2.0.tgz --username='$oauthtoken' --password=<YOUR API KEY>

helm upgrade --install \
  --namespace embedding-nim \
  --set image.repository=nvcr.io/nim/snowflake/arctic-embed-l \
  --set image.tag=1.0.1 \
  --set persistence.class="local-nfs" \
  nemo-embedder \
  text-embedding-nim-1.2.0.tgz

Deploying Mistral 7B#

Create a values files for the resource requirements of the 7B Model:

# values-mistral.yaml
resources:
  limits:
    ephemeral-storage: 28Gi
    nvidia.com/gpu: 1
    memory: 32Gi
    cpu: "16000m"
  requests:
    ephemeral-storage: 28Gi
    nvidia.com/gpu: 1
    memory: 16Gi
    cpu: "4000m"

Then deploy the model:

helm fetch https://helm.ngc.nvidia.com/nim/nvidia/charts/text-embedding-nim-1.2.0.tgz --username='$oauthtoken' --password=<YOUR API KEY>

helm upgrade --install \
  --namespace embedding-nim \
  -f values-mistral.yaml \
  --set image.repository=nvcr.io/nim/nvidia/nv-embedqa-mistral-7b-v2 \
  --set image.tag=1.0.1 \
  --set persistence.class="local-nfs" \
  nemo-embedder \
  text-embedding-nim-1.2.0.tgz

Running Inference#

In the previous example the API endpoint is exposed on port 8000 through the Kubernetes service of the default type with no ingress, since authentication is not handled by the NIM itself. The following commands assume the llama-3.2-nv-embedqa-1b-v2 model was deployed.

Adjust the “model” value in the request JSON body to use a different model.

Use the following command to port-forward the service to your local machine to test inference.

kubectl port-forward -n embedding-nim service/nemo-embedder-nvidia-nim-llama-32-nv-embedqa-1b-v2 8000:8000

Then try a request:

curl -X 'POST' \
  'http://localhost:8000/v1/embeddings' \
  -H 'accept: application/json' \
  -H 'Content-Type: application/json' \
  -d '{
    "input": "hello world",
    "model": "nvidia/llama-3.2-nv-embedqa-1b-v2",
    "input_type": "passage"
  }'

Viewing Log Messages#

Use the following command to view the container log messages in the docker logs.

kubectl logs -f nvidia-nim-llama-32-nv-embedqa-1b-v2-0
On this page