Skip to main content

AI on Edge Flagship Accelerator

Build Status License: MIT Open in Dev Containers

What You'll Find Here​

🎯 For Users​

Ready to achieve rapid edge-ai deployments? Start with our General User Guide to deploy existing blueprints to Azure in 30-60 minutes.

🏗️ For Blueprint Developers​

Creating new deployment scenarios? Start your process with the Blueprint Developer Guide to learn how to combine components into custom solutions that achieve your business goals.

⚙️ For Feature Developers​

Contributing new capabilities? The Feature Developer Guide empowers you with component development, testing, and contribution workflows to achieve impactful contributions.

Key Features​

  • Production-Ready: Battle-tested Infrastructure as Code that empowers organizations to achieve repeatedly deployable and reliable edge AI scenarios
  • Modular Design: Reusable components that enable teams to achieve custom solutions tailored to their unique business requirements
  • Multiple Frameworks: Support for both Terraform and Bicep for diverse technical requirements
  • AI-Assisted Development: Optimized for GitHub Copilot and AI-powered development workflows that accelerate team productivity
  • Comprehensive Testing: Automated validation and testing that ensures global-scale reliability for every deployment
  • Edge-Focused: Purpose-built capabilities that empower organizations worldwide to achieve edge AI computing workload success

Repository Overview​

How to Use This Repository​

flowchart TD
    Start[I want to implement<br/>Edge AI solutions]

    %% User approach choices
    Quick[Quick Deploy<br/>Use existing blueprints]
    Custom[Custom Solution<br/>Build with components]

    %% Repository structure navigation
    Blueprints[blueprints/<br/>Ready-to-deploy<br/>solutions]
    Components[src/<br/>Reusable<br/>components]
    Docs[docs/<br/>Documentation<br/>& guides]

    %% Common implementation scenarios
    PM[Predictive<br/>Maintenance]
    OPM[Performance<br/>Monitoring]
    QO[Process<br/>Optimization]

    %% Primary user flow
    Start --> Quick
    Start --> Custom
    Start --> Docs

    Quick --> Blueprints
    Custom --> Components
    Docs --> Components
    Components --> Blueprints

    %% Application to business scenarios
    Blueprints --> PM
    Blueprints --> OPM
    Blueprints --> QO

    %% Comprehensive color scheme for repository workflow
    style Start fill:#e1f5fe,stroke:#01579b,stroke-width:3px
    style Quick fill:#e8f5e8,stroke:#2e7d32,stroke-width:2px
    style Custom fill:#fff3e0,stroke:#e65100,stroke-width:2px
    style Blueprints fill:#e8f5e8,stroke:#2e7d32,stroke-width:2px
    style Components fill:#fff3e0,stroke:#e65100,stroke-width:2px
    style Docs fill:#f3e5f5,stroke:#7b1fa2,stroke-width:2px
    style PM fill:#cffafe,stroke:#059669,stroke-width:2px
    style OPM fill:#cffafe,stroke:#059669,stroke-width:2px
    style QO fill:#cffafe,stroke:#059669,stroke-width:2px

🎯 How Blueprints Build Edge Solutions​

Blueprints are pre-configured compositions that combine Cloud Foundation, Edge Infrastructure, IoT Platform, and Observability components to deliver Industrial Automation, AI Workloads, and System Reliability.

🏗️ Nine Blueprint Types​

  • Full Single Node → Complete solution with all components for comprehensive edge deployment
  • Full Multi Node → Enhanced distributed edge computing with load balancing and redundancy
  • Full Arc Multi Node → Hybrid cloud + edge with AKS and multiple edge nodes
  • Minimal Single Node → Core components only for resource-optimized deployment
  • Partial Single Node → Partially configured edge solution for specific use cases
  • Edge IoT Only → Add Azure IoT Operations to existing infrastructure
  • Cloud Only → Hosting-ready cloud infrastructure for edge workloads
  • CNCF Cluster Script → Automated deployment scripts for Kubernetes clusters
  • Fabric → Advanced analytics and data platform for edge-to-cloud insights

☁️ Cloud Foundation provides the supporting infrastructure​

  • Resource Management: Resource groups, organization, governance
  • Security & Identity: Authentication, RBAC, Key Vault, certificates
  • Data Services: Data lakes, storage accounts, time-series databases
  • Messaging Services: Event Grid, Event Hubs, Service Bus

🖥️ Edge Infrastructure delivers the compute platform​

  • VM Hosting: Virtual machines for edge hosting and management
  • Kubernetes Cluster: K3s with Arc-enabled management and orchestration
  • Networking: VNets, security groups, private endpoints

🏭 IoT Platform enables industrial connectivity​

  • MQTT Broker: Secure messaging and communication hub
  • Data Processing: Real-time stream processing and analytics
  • Protocols: Industrial protocol translation and device integration

🔧 Device Management handles asset connectivity​

  • OPC UA Assets: Industrial device integration and asset modeling
  • Asset Discovery: Automatic detection and onboarding of devices

📊 Observability ensures system health​

  • Cloud Monitoring: Application Insights, Log Analytics, dashboards
  • Edge Monitoring: Local monitoring, health checks, performance metrics

🔧 Analytics Platform provides advanced data capabilities​

  • Real-time Analytics: Stream processing and live data analysis
  • AI/ML Services: Machine learning model deployment and inference
  • Business Intelligence: Reporting, dashboards, and data visualization

🔗 Integration Services enable data flow and connectivity​

  • Data Pipelines: ETL/ELT processes and data transformation
  • Event Streaming: Real-time event processing and routing
  • API Management: Service exposure and integration management

⚙️ Automation Tools streamline deployment and management​

  • Deployment Scripts: Automated infrastructure provisioning
  • Configuration Management: Consistent system configuration and updates

Quick Start​

  1. Choose your path from our Getting Started Guides
  2. Set up your environment with our Dev Container
  3. Deploy a blueprint from our Blueprint Catalog
  4. Explore components in our Component Library

Note on Telemetry: If you wish to opt-out of sending telemetry data to Microsoft when deploying Azure resources with Terraform, you can set the environment variable ARM_DISABLE_TERRAFORM_PARTNER_ID=true before running any terraform commands.

Community and Support​

Ready to get started? Head to our Getting Started Guides and choose the path that matches your role!

Responsible AI​

Microsoft encourages customers to review its Responsible AI Standard when developing AI-enabled systems to ensure ethical, safe, and inclusive AI practices. Learn more at Microsoft's Responsible AI.

📄 Legal​

This project is licensed under the MIT License.

Security: See SECURITY.md for security policy and reporting vulnerabilities.

Trademark Notice​

This project may contain trademarks or logos for projects, products, or services. Authorized use of Microsoft trademarks or logos is subject to and must follow Microsoft's Trademark & Brand Guidelines. Use of Microsoft trademarks or logos in modified versions of this project must not cause confusion or imply Microsoft sponsorship. Any use of third-party trademarks or logos are subject to those third-party's policies.

🤖 Crafted with precision by ✨Copilot following brilliant human instruction, then carefully refined by our team of discerning human reviewers.