From Training to Real-Time Inference: How to Solve Computer Vision Challenges in Healthcare Learn More >

Wallaroo Community: A Quick Guide to Installing and Opening Your Licensed Account

Wallaroo community edition quick start quide | Wallaroo.AI
https://player.vimeo.com/video/716517936?h=aa7d5e2030&app_id=122963

Ready to install your Wallaroo Community Edition? If you are looking for a simple and efficient way to deploy real-time applications and machine learning models, look no further than Wallaroo Community Edition, a cloud-based AI data processing platform where you can manage your workspaces, models, and pipelines and collaborate with other users.

In this guide, we are going to take you through the process of installing the software in your Kubernetes environment to accessing your Wallaroo Community dashboard.

Prerequisites Before Starting

Before you get started on the steps, here are some of the requirements you must meet:

  • You need a Microsoft Azure account.
  • Connect your Microsoft account when installing the az command line program or Azure CLI.
  • Requires kubectl and kots installed on your system for the setup.
  • The Kubernetes cluster requires each node to have 8 CPU cores minimum, with a minimum of 4 nodes and 16 GB RAM.
  • The default container driver must be containerd and needs to be specified, during the process.

Please Note: The following information has been validated only as of 6/28/2022 and the recorded procedures in this blog may not reflect current practices. If you have any issues executing these procedures with similar results, please visit our documentation site for our most recent code and suggestions.

Installing Wallaroo Community Azure Kubernetes Environment

During this installation process, we are using one node each for setting up postgres, main pool, and the two engine pools, along with the most suitable machine types. Depending on the requirements, you can change node numbers per nodepool and machine types as you want.


Step 1: Azure Resource Group

To create your Azure Resource Group, you need to name the resource group and the cluster installation region. So for resource group “wallarooCEGroup” and region east US, the command would be:

az group create –name wallarooCEGroup –location eastus

You can further set the default Resource Group to the one you just created. It will create an automatic selection for the group for az aks list commands. In this case, it’s:

az configure  –defaults group=wallarooCEGroup

Step 2: Azure Container Registry

Now, you can move on to creating ACR (Azure Container Registry) that is in charge of the container images for Kubernetes involving services. For this, you will need the resource group name, the new ACR name, and the sku. Here we will be using the Basic sku, and the command would be:

az acr create –n wallarooCEAcr –g wallarooCEGroup –sku Basic

You may need to wait for a few minutes for it to be complete. After that, you can start with the next step, which is creating the Kubernetes service.

Step 3: Kubernetes Service

The Azure Kubernetes service will host your Wallaroo instances. We will use the nodepool mainpool for this cluster, which the Kubernetes service creates. The command for supporting Wallaroo in the Azure Kubernetes environment would look like the following:

az aks create \

–resource-group wallarooCEGroup \

–name wallarooCEAKS \

–node-count 4 \

–vm-set-type VirtualMachineScalesSets \

–load-balancer-sku standard \

–node-vm-size Standard_D8s_v4

–nodepool –name wallarooCEPool \

–nodepool –name mainpool \

–attach-acr wallarooceacr

At the end of this command, you will see the specified Kubernetes version to use. The recent Kubernetes version for Wallaroo is 1.21.9

Step 4: Setting Nodepools

For Wallaroo Community, you can only use static nodepools. If we set the mainpool, postgres, enginelb, and engine as the four nodepools, we will have to specify the virtual machine size, along with the nodepool name, and the assigned cluster for the nodepools.

In case of mainpool and postgres nodepools, for the virtual machine size, we can use:

VM_SIZE=Standard_D8s_v4

It will provide 8 CPU cores and 16 GB of RAM. As for the machine type for enginelb and engine nodepools, the command that supports the Wallaroo Community cluster would be:

VM_SIZE=Standard_F8s_v2

Step 5: Retrieving Kubernetes Credentials

Once you are done setting up the nodepools, you can start retrieving your Kubernetes credentials to your local system. If you have been following the process in a correct way, you should know that our resource group is called wallarooCEGroup, and our cluster is wallarooCEAKS. With that, the command to retrieve credentials would be:

az aks get-credentials –resource-group wallarooCEGroup –name wallarooCEAKS

You can verify if your Kubernetes cluster is ready or not with the command:

kubectl get nodes

Step 6: Installing Wallaroo Community

With that step, now you are ready to install the Wallaroo Community to your Kubernetes environment. Begin with a namespace command where your Wallaroo Community is installed.

kubectl create namespace wallaroo

After that, you can enter the following command to finish installing Wallaroo Community. It will download all the important containers in your Kubernetes environment.

kubectl kots install wallaroo/ce –namespace wallaroo

You need to apply your default administrative password before completing the installation process.

Access Your Dashboard

After installing Wallaroo Community, you can visit http://localhost:8800 to go to your Wallaroo Community dashboard. You need to log in with your password and upload your Wallaroo Community license file.

It takes you to a configuration page where you can choose the most appropriate options for you before clicking the “continue” button. After performing the Preflight checks to ensure your Kubernetes environment is ready, you can access your dashboard.

You can see your main prefix and DNS suffix, making up your Wallaroo Community dashboard URL on your dashboard profile. Wallaroo will email you with a temporary password on your first sign-in, which you can change for a permanent one.

Make sure to replace all the example command templates according to your organization’s needs. You can also check the Wallaroo documentation site for further details and assistance.  

Table of Contents

Topics

Share

Related Blog Posts

Get Your AI Models Into Production, Fast.

Unblock your AI team with the easiest, fastest, and most flexible way to deploy AI without complexity or compromise. 

Keep up to date with the latest ML production news sign up for the Wallaroo.AI newsletter

Platform
Platform Learn how our unified platform enables ML deployment, serving, observability and optimization
Technology Get a deeper dive into the unique technology behind our ML production platform
Solutions
Solutions See how our unified ML platform supports any model for any use case
Computer Vision (AI) Run even complex models in constrained environments, with hundreds or thousands of endpoints
Resources