This document will introduce you to all you need to know to get started with version 2 of Charmed MLflow.
We are assuming that you are running this tutorial on a local machine supporting snaps.
- Install and prepare MicroK8s
- Install Juju
- Create a model called
- Deploy mlflow bundle
- Access MLflow
- Object storage credentials
Install and prepare MicroK8sLet’s install MicroK8s. MicroK8s is installed from a snap package. The published snap maintains different
channelsfor different releases of Kubernetes.
sudo snap install microk8s --classic --channel=1.24/stable
For MicroK8s to work without having to use
sudo for every command, it creates a group called
microk8s. To make it more convenient to run commands, you will add the current user to this group:
sudo usermod -a -G microk8s $USER newgrp microk8s
It is also useful to make sure the user has the proper access and ownership of any
kubectl configuration files:
sudo chown -f -R $USER ~/.kube
Enable the following Microk8s add-ons to configure your Kubernetes cluster with extra services needed to run Charmed Kubeflow.
microk8s enable dns storage ingress metallb:10.64.140.43-10.64.140.49
Here, we added a DNS service, so the applications can find each other, storage, an ingress controller so we can access Kubeflow components and the MetalLB load balancer application. You can see that we added some detail when enabling MetalLB, in this case the address pool to use.
See More : Microk8s | How to use addons
We’ve now installed and configured MicroK8s. It will start running automatically, but can take 5 minutes or so before it’s ready for action.
Run the following command to tell MicroK8s to report its status to us when it’s ready:
microk8s status --wait-ready
Be patient - this command may not return straight away. The
--wait-ready flag tells MicroK8s to wait for the Kubernetes services to initialise before returning. Once MicroK8s is ready, you will see something like the following output:
microk8s is running
Below this there will be a bunch of other information about the cluster.
Great, we have now installed and configured MicroK8s, and it’s running and ready!
Install JujuJuju is an operation Lifecycle manager (OLM) for clouds, bare metal or Kubernetes. We will be using it to deploy and manage the components which make up Kubeflow.
To install Juju from snap, run this command:
sudo snap install juju --classic --channel=2.9/stable
Now, run the following command to deploy a Juju controller to the Kubernetes we set up with MicroK8s:
juju bootstrap microk8s
Sit tight while the command completes! The controller may take a minute or two to deploy.
The controller is Juju’s agent, running on Kubernetes, which can be used to deploy and control the components of Kubeflow.
Next, we’ll need to add a model for Kubeflow to the controller. Run the following command to add a model called
juju add-model kubeflow
The controller can work with different
models, which map 1:1 to namespaces in Kubernetes. In this case, the model name must be
kubeflow, due to an assumption made in the upstream Kubeflow Dashboard code.
Great job: Juju has now been installed and configured for Kubeflow!
Deploy MLflow bundle
Let’s now use Juju to deploy Charmed MLflow. Run the following command:
juju deploy mlflow --channel=2.1/edge --trust
This deploys the latest edge version of MLflow with MinIO as object storage and MySQL as metadata store.
To access MLflow, visit the following URL in your web browser:
This will take you to the MLflow UI.
Note: by default Charmed MLflow creats a nodeport on port 31380 where you can access the MLflow UI.
That’s it! Charmed MLflow has been deployed locally with microk8s and Juju. You can now start using MLflow.
Reference: Object storage credentialsTo use mlflow you need to have credentials to the object storage. The aforementioned bundle comes with minio. To get the minio credentials run the following command:
juju run-action mlflow-server/0 get-minio-credentials --wait
This action will output