Step 3: Design

📊 Progress: Step 3 of 4 ⏱️ Estimated Time: 5-6 hours

Executive Summary

This step represents the core architecture work for the AI Hub solution. You’ll create a comprehensive technical design that addresses IFS’s requirements while following Azure best practices for secure, scalable AI platforms. Your design will establish a central foundation for accelerating AI innovation across the organization.

Home > AI Hub Challenge > Step 3 - Design

This section is part of the IFS AI Hub Challenge. Here, you’ll design the end-to-end architecture for the AI Hub, ensuring alignment with Azure best practices and reference architectures.

Table of Contents

Note: Generated using Jekyll’s automatic table of contents feature

flowchart LR
    A[🚀 Start] --> B[📝 Step 1 Scenario]
    B --> C[📋 Step 2 Requirements]
    C --> D[🏗️ Step 3 Design]
    D -->|Current| E[📊 Step 4 Presentation]
    style D fill:#90EE90,stroke:#333,stroke-width:2px

🎯 Objective

[!NOTE] This step is critical for the success of the AI Hub Challenge. A well-designed architecture ensures that the AI Hub can scale securely to support the organization’s AI initiatives.

Define the architecture, controls, and Azure services needed for a robust, secure, and scalable AI Hub solution.

🔝 Back to Top

📝 Activities

  • Review your findings from Steps 1 and 2.
  • As a team, design and document:
    1. Architecture: Draw a high-level architecture diagram showing the AI Hub architecture.
    2. Service Selection: List Azure services that will be part of the AI Hub. 3. Network Security: Detail the network design including VNets, subnets, NSGs, and private endpoints.

    [!IMPORTANT] Ensure your network design includes private endpoints for all Azure AI services to maintain data security and comply with data residency requirements. This is particularly critical for AI services that process sensitive information.

    1. Identity & Access: Define the identity model, service principals, managed identities, and RBAC.
    2. Cost Controls: Design the approach for cost tagging, budgets, and alerts.
    3. Monitoring & Operations: Design the monitoring strategy, logs, and alerts.
    4. Resource Management: Define the resource group structure, tagging strategy, and governance controls.

🔝 Back to Top

Guidance

[!TIP] Start with a high-level architecture diagram that shows the main components of your AI Hub, then drill down into specific areas such as networking, security, and service deployment.

Best Practice: Reference the Azure OpenAI baseline Landing Zone reference architecture and apply these principles to your broader AI Hub design.

Sample AI Hub Architecture Diagram 

flowchart TB
    subgraph "AI Hub Platform - Shared Services"
        direction TB
        subgraph "AI Core Services"
            AOAI[Azure OpenAI Service]
            ACS[Azure AI Search]
            ACV[Azure Computer Vision]
            ADS[Azure Document Intelligence]
            AST[Azure Speech Services]
        end
        
        subgraph "Data Services"
            STRG[Azure Storage]
            COSMOS[Azure Cosmos DB]
            SQLDB[Azure SQL Database]
            DL[Azure Data Lake]
        end
        
        subgraph "Platform Services"
            APIM[API Management]
            AKV[Azure Key Vault]
            AADB2C[Azure AD B2C]
            RBAC[Role-Based Access]
        end
        
        subgraph "DevOps & Governance"
            ADO[Azure DevOps]
            ACR[Azure Container Registry]
            PAR[Policy & Regulatory Controls]
        end
        
        subgraph "Monitoring & Operations"
            LOG[Log Analytics]
            APPINS[Application Insights]
            MONANOM[Anomaly Detection]
        end
    end
    
    subgraph "Consumer Layer"
        APP1[Business App 1]
        APP2[Business App 2]
        APP3[Business App 3]
        PORTAL[AI Services Portal]
    end
    
    APP1 --> APIM
    APP2 --> APIM
    APP3 --> APIM
    PORTAL --> APIM
    
    APIM --> AOAI
    APIM --> ACS
    APIM --> ACV
    APIM --> ADS
    APIM --> AST
    
    AOAI --> AKV
    ACS --> AKV
    ACV --> AKV
    ADS --> AKV
    AST --> AKV
    
    AOAI --> STRG
    ACS --> STRG
    ACV --> STRG
    ADS --> STRG
    AST --> STRG
    
    APIM --> AADB2C
    AADB2C --> RBAC
    
    AOAI -.-> LOG
    ACS -.-> LOG
    ACV -.-> LOG
    ADS -.-> LOG
    AST -.-> LOG
    APIM -.-> LOG
    
    LOG --> APPINS
    APPINS --> MONANOM
    
    classDef aiservices fill:#0078D4,color:white,stroke:#0078D4,stroke-width:1px;
    classDef dataservices fill:#773BD9,color:white,stroke:#773BD9,stroke-width:1px;
    classDef platformservices fill:#107C10,color:white,stroke:#107C10,stroke-width:1px;
    classDef devops fill:#FFB900,color:black,stroke:#FFB900,stroke-width:1px;
    classDef monitoring fill:#F25022,color:white,stroke:#F25022,stroke-width:1px;
    classDef consumer fill:#50E6FF,color:black,stroke:#0078D4,stroke-width:1px;
    
    class AOAI,ACS,ACV,ADS,AST aiservices;
    class STRG,COSMOS,SQLDB,DL dataservices;
    class APIM,AKV,AADB2C,RBAC platformservices;
    class ADO,ACR,PAR devops;
    class LOG,APPINS,MONANOM monitoring;
    class APP1,APP2,APP3,PORTAL consumer;

🔝 Back to Top

Success Criteria ✅

By the end of this step, you should have:

  • ✓ A comprehensive AI Hub architecture diagram
  • ✓ Clear rationale for Azure service selections
  • ✓ Documentation of network, identity, cost, and operational controls
  • ✓ A governance model that meets IFS requirements

To successfully complete this step, your design should provide a complete technical blueprint for implementing the AI Hub that addresses all requirements identified in the previous step.

🔝 Back to Top

[!WARNING] Before moving to Step 4, ensure your architecture design addresses all the requirements identified in Step 2, including security, governance, and operational considerations.