deployment.md

Deployment Guide

Prerequisites

If you do not have an Azure Subscription, you can sign up for a free Azure account and create an Azure Subscription.

To deploy this solution successfully, your Azure account must have the following permissions on the targeted Azure Subscription:

• Microsoft.Authorization/roleAssignments/write permissions at the subscription scope. (typically included if you have Role Based Access Control Administrator, User Access Administrator, or Owner role)
• Microsoft.Resources/deployments/write permissions at the subscription scope.

For more details on Azure account setup, see the Azure account setup guide.

Additional Requirements

• VI Arc Extension Approval - Submit your subscription for approval via this form before deploying
• GPU Quota - Sufficient GPU quota for your chosen VM sizes in your target region. See the quota check guide for details
• Required CLI Tools:
  • Azure CLI (v2.x+)
  • kubectl
  • Helm 3
  • Azure Developer CLI (azd) (v1.18.0+)

Region Availability

Check the Azure Products by Region page and select a region where the following services are available:

• Azure Kubernetes Service (AKS)
• Azure Arc
• Azure Video Indexer
• Azure AI Foundry (if deploying AI agents)
• GPU-enabled VM SKUs (e.g., NCasv3, NCadsA100v4 families)

Important Note for PowerShell Users

If you encounter issues running PowerShell scripts due to execution policy, you can temporarily adjust it:

Set-ExecutionPolicy -Scope Process -ExecutionPolicy Bypass

Deployment Options & Steps

Pick from the options below to see step-by-step instructions for GitHub Codespaces, VS Code Dev Containers, or local environments.

Deploy in GitHub Codespaces

GitHub Codespaces

You can run this template virtually by using GitHub Codespaces. The button will open a web-based VS Code instance in your browser:

1. Open the template (this may take several minutes):

   

2. Open a terminal window

3. Continue with the deploying steps

Deploy in VS Code Dev Containers

VS Code Dev Containers

A related option is VS Code Dev Containers, which will open the project in your local VS Code using the Dev Containers extension:

1. Start Docker Desktop (install it if not already installed: Docker Desktop)

2. Open the project:

   

3. In the VS Code window that opens, once the project files show up (this may take several minutes), open a terminal window

4. Continue with the deploying steps

Deploy in your local environment

Local Environment

If you're not using one of the above options, then you'll need to:

1. Make sure the following tools are installed:

   • Azure Developer CLI (azd) (v1.18.0+)
   • Azure CLI (v2.x+)
   • kubectl
   • Helm 3
   • Git
   • [Windows Only] PowerShell - Make sure pwsh.exe is added to PATH

2. Clone the repository:

   azd init -t Video-Agents-Foundry-Solution

3. Open the project folder in your terminal or editor

4. Continue with the deploying steps

Configurable Deployment Settings

When you start a deployment, most parameters will have default values. You can change the following settings:

Setting	Description	Default
Azure Region	Select a region with GPU quota	eastus2
Kubernetes Version	AKS cluster Kubernetes version	1.32
System VM Size	System node pool VM size	Standard_D4a_v4
Workload VM Size	Workload node pool VM size	Standard_D32a_v4
DeepStream GPU VM Size	DeepStream GPU node pool	Standard_NC24ads_A100_v4
Inference GPU VM Size	Inference GPU node pool	Standard_NC24ads_A100_v4
Create Foundry Project	Deploy AI Foundry project	true
AI Model Name	OpenAI model for agents	gpt-5.2
AI Model Version	Model version	2025-12-11

For a detailed description of customizable fields, view the deployment customization guide.

Deploying with AZD

Once you've opened the project in Codespaces, Dev Containers, or locally, deploy to Azure:

1. (Optional) Customize your deployment settings. See the deployment customization guide for details.

2. Run the deployment:

   azd up

3. You will be prompted to:

   • Provide an azd environment name (e.g., "vi-agents-solution")
   • Select a subscription from your Azure account
   • Select a location with sufficient GPU quota
   • Choose VM sizes for each AKS node pool (interactive menu with live quota checking)

   Note: The preprovision hook validates prerequisites including Azure CLI authentication, required CLI tools, Azure resource provider registration, and GPU quota availability.

4. The deployment process will:

   • Provision Azure resources via Bicep (AKS cluster, Video Indexer account, storage, managed identity, and optionally AI Foundry)
   • Connect AKS to Azure Arc
   • Install NVIDIA GPU Operator via Helm
   • Deploy Video Indexer Arc Extension
   • Run a health check dashboard showing pod status across namespaces

   This process typically takes 15-25 minutes. After azd up completes, allow an additional 20-35 minutes for GPU drivers and VI extension pods to fully initialize (45-60 minutes total).

5. After deployment completes, the health dashboard will show the status of all components:

   • gpu-operator namespace (NVIDIA GPU Operator)
   • video-indexer namespace (VI Arc Extension)
   • app-routing-system namespace
   • azure-arc namespace
   • cert-manager namespace

6. View information about your deployment:

   azd show

Resource Cleanup

When you no longer need the deployed resources, clean up by running:

azd down

This will delete the resource group and all associated Azure resources.

GitHub Actions CI/CD

This repo includes a GitHub Actions workflow (.github/workflows/azure-dev.yml) for automated deployments. To set it up:

1. Configure OpenID Connect (OIDC) federated credentials for your Azure service principal
2. Set the required GitHub repository variables (see the workflow file for the full list)
3. Set the AZD_INITIAL_ENVIRONMENT_CONFIG secret

The workflow triggers on push to main or manual dispatch, and runs azd provision followed by azd deploy.

Next Steps

• Review the troubleshooting guide if you encounter issues
• Check deployment customization for advanced configuration
• Explore observability features for monitoring and tracing