🏗️ Architecture Documentation¶

🏠 Home > 📚 Docs > 🏗️ Architecture

Last Updated: 2025-01-21 | Version: 1.0.0

📑 Table of Contents¶

• 🎯 Overview
• 🏛️ High-Level Architecture
• 🥉🥈🥇 Medallion Architecture
• 🥉 Bronze Layer (Raw Data)
• 🥈 Silver Layer (Cleansed Data)
• 🥇 Gold Layer (Business Ready)
• ⚡ Real-Time Intelligence Architecture
• 📊 Data Governance
• 🔐 Security Architecture
• 📈 Capacity Planning
• 🔄 Disaster Recovery
• 📡 Monitoring & Alerting
• 🛠️ Technology Decisions

🎯 Overview¶

This document describes the architecture of the Microsoft Fabric Casino/Gaming POC environment. The solution implements a modern data lakehouse architecture using the medallion pattern (Bronze/Silver/Gold) with real-time analytics capabilities.

📝 Note: This architecture is designed for a Proof of Concept (POC) environment. Production implementations may require additional security controls, compliance certifications, and capacity planning.

🏛️ High-Level Architecture¶

Microsoft Fabric Platform Architecture¶

Microsoft Fabric provides a unified SaaS experience that integrates all data and analytics workloads. The diagram below shows how Fabric's core components work together:

Source: Microsoft Fabric Overview

OneLake: The Foundation¶

OneLake serves as the single, unified data lake for your entire organization. All Fabric workloads automatically store data in OneLake using the Delta Lake format:

Source: OneLake Overview

Casino/Gaming POC Architecture¶

flowchart TB
    subgraph Sources["🎰 Data Sources"]
        SAS["🎰 Slot Machines<br/>SAS Protocol"]
        TG["🃏 Table Games<br/>RFID/Terminals"]
        LMS["👤 Loyalty System"]
        CAGE["💰 Cage Operations"]
        SEC["🔒 Security/Surveillance"]
        COMP["📋 Compliance Systems"]
    end

    subgraph Ingestion["📥 Ingestion Layer"]
        ES["⚡ Eventstreams<br/>Real-Time"]
        DF["📊 Dataflows Gen2<br/>Batch"]
        PIPE["🔧 Data Pipelines"]
    end

    subgraph Fabric["☁️ Microsoft Fabric"]
        subgraph Bronze["🥉 Bronze Layer"]
            B_SLOT[bronze_slot_telemetry]
            B_TABLE[bronze_table_games]
            B_PLAYER[bronze_player_profile]
            B_FIN[bronze_financial_txn]
            B_SEC[bronze_security_events]
        end

        subgraph Silver["🥈 Silver Layer"]
            S_SLOT[silver_slot_cleansed]
            S_TABLE[silver_table_enriched]
            S_PLAYER[silver_player_master]
            S_FIN[silver_financial_reconciled]
            S_SEC[silver_security_enriched]
        end

        subgraph Gold["🥇 Gold Layer"]
            G_SLOT[gold_slot_performance]
            G_TABLE[gold_table_analytics]
            G_PLAYER[gold_player_360]
            G_FIN[gold_financial_summary]
            G_SEC[gold_compliance_reporting]
        end

        subgraph Analytics["📈 Analytics"]
            DL["🔗 Direct Lake<br/>Semantic Model"]
            PBI["📊 Power BI<br/>Reports"]
            RTD["⏱️ Real-Time<br/>Dashboards"]
        end
    end

    subgraph Governance["🛡️ Governance"]
        PV["Microsoft Purview"]
    end

    Sources --> Ingestion
    Ingestion --> Bronze
    Bronze --> Silver
    Silver --> Gold
    Gold --> Analytics
    Fabric --> Governance

    style Bronze fill:#CD7F32,color:#000
    style Silver fill:#C0C0C0,color:#000
    style Gold fill:#FFD700,color:#000

Component Summary¶

🥉🥈🥇 Medallion Architecture¶

The medallion architecture provides a structured approach to data refinement. This pattern is a recommended best practice for organizing data in a lakehouse:

Source: Implement medallion lakehouse architecture in Fabric

Medallion Layer Flow¶

flowchart LR
    subgraph B["🥉 BRONZE<br/>Raw Data"]
        B1["Schema-on-read"]
        B2["Append-only"]
        B3["Full fidelity"]
    end

    subgraph S["🥈 SILVER<br/>Cleansed Data"]
        S1["Schema enforced"]
        S2["Data quality"]
        S3["Deduplication"]
    end

    subgraph G["🥇 GOLD<br/>Business Ready"]
        G1["Aggregated"]
        G2["Star schema"]
        G3["Direct Lake optimized"]
    end

    B --> S --> G

    style B fill:#CD7F32,color:#000
    style S fill:#C0C0C0,color:#000
    style G fill:#FFD700,color:#000

🥉 Bronze Layer (Raw Data)¶

Purpose: Land raw data with minimal transformation for auditability and reprocessing.

🥈 Silver Layer (Cleansed Data)¶

Purpose: Validated, cleansed, and enriched data with enforced schema.

💡 Pro Tip: Implement data quality checks at the Silver layer to catch issues early. Use Great Expectations or Delta Lake constraints for automated validation.

📝 Note: Key Characteristics: - Schema enforcement (Delta Lake) - Data quality rules applied - Referential integrity checked - Business keys established

🥇 Gold Layer (Business Ready)¶

Purpose: Aggregated, business-oriented views optimized for analytics.

💡 Pro Tip: Design Gold tables with Power BI consumption in mind. Use proper partitioning and avoid wide tables to optimize Direct Lake performance.

📝 Note: Key Characteristics: - Star schema design - Pre-aggregated metrics - Direct Lake optimized - Incremental refresh enabled

⚡ Real-Time Intelligence Architecture¶

Microsoft Fabric Real-Time Intelligence provides end-to-end streaming analytics capabilities. The architecture below shows how real-time data flows through the platform:

Source: Real-Time Intelligence Overview

Casino Floor Real-Time Architecture¶

flowchart TB
    subgraph Sources["📡 Real-Time Sources"]
        SLOT["🎰 Slot Machines"]
        TABLE["🃏 Table Games"]
        SEC["🔒 Security"]
    end

    subgraph Streaming["⚡ Streaming"]
        ES["Eventstream"]
    end

    subgraph RealTime["📊 Real-Time Intelligence"]
        EH_DB[("Eventhouse<br/>KQL Database")]
        KQL["KQL Queries"]
        ALERT["🔔 Alerts"]
    end

    subgraph Dashboard["📺 Dashboards"]
        RT_DASH["Real-Time Dashboard"]
        FLOOR["Floor Monitor"]
    end

    Sources --> Streaming --> RealTime --> Dashboard
    ALERT -->|"Jackpot > $10K"| FLOOR
    ALERT -->|"Machine Down"| FLOOR

📊 Data Governance¶

Microsoft Purview provides unified data governance across your entire data estate. The Purview hub in Fabric gives you a central place to manage data discovery, lineage, and access policies.

Source: Use Microsoft Purview hub in Fabric

Purview Integration¶

flowchart LR
    subgraph Purview["🛡️ Microsoft Purview"]
        CAT["📚 Data Catalog"]
        LIN["🔗 Data Lineage"]
        POL["🔐 Access Policies"]
    end

    subgraph Features["Features"]
        F1["Glossary terms"]
        F2["Classifications"]
        F3["Ownership"]
        F4["Impact analysis"]
        F5["Sensitivity labels"]
        F6["Row-level security"]
    end

    Purview --> Features

Data Classification¶

⚠️ Warning: PII data must be handled according to gaming regulations (NIGC MICS, state regulations) and may be subject to audit. Never store unencrypted SSN or full card numbers in the Gold layer.

💡 Pro Tip: Use dynamic data masking or column-level encryption for sensitive data. Purview can automatically discover and classify PII fields.

🔐 Security Architecture¶

Network Architecture¶

flowchart TB
    subgraph VNet["🌐 Virtual Network"]
        subgraph Fabric["Fabric Subnet<br/>10.0.1.0/24"]
            F1["Fabric Workspace"]
        end
        subgraph PE["Private Endpoint Subnet<br/>10.0.2.0/24"]
            P1["Storage PE"]
            P2["Key Vault PE"]
            P3["Purview PE"]
        end
        subgraph Mgmt["Management Subnet<br/>10.0.3.0/24"]
            M1["Admin Access"]
        end
    end

Identity & Access Controls¶

📋 Prerequisites: For production deployments, implement private endpoints and disable public network access to OneLake and Key Vault.

📈 Capacity Planning¶

🔄 Disaster Recovery¶

RPO/RTO Targets¶

💡 Pro Tip: Test your disaster recovery procedures quarterly. Use Delta Lake time travel to practice point-in-time recovery scenarios.

📡 Monitoring & Alerting¶

Key Metrics¶

Alert Thresholds¶

🛠️ Technology Decisions¶

Direct Lake Mode¶

Direct Lake is the recommended connectivity mode for Power BI in Fabric. It provides the performance of import mode with the freshness of DirectQuery:

Source: Direct Lake Overview

📚 Related Documentation¶

⬆️ Back to top

📖 Documentation maintained by: Microsoft Fabric POC Team 🔗 Repository: Suppercharge_Microsoft_Fabric