Home > Docs > Features > Real-Time Hub

📡 Real-Time Hub — Event Discovery and Sharing¶

Last Updated: 2026-04-21 | Version: 1.0.0

📑 Table of Contents¶

• 🎯 Overview
• 🏗️ Architecture
• ⚙️ Event Sources
• 🔄 Streams and Topics
• 🎰 Casino Implementation
• 🏛️ Federal Implementation
• ⚠️ Limitations
• 📚 References

🎯 Overview¶

Real-Time Hub is a centralized event catalog in Microsoft Fabric that makes all streaming data discoverable, shareable, and consumable across the organization. Instead of creating point-to-point Eventstream connections, teams publish events to the Real-Time Hub where other teams can browse, subscribe, and build derived streams.

Think of Real-Time Hub as a "marketplace" for streaming data — just as OneLake Catalog indexes batch data, Real-Time Hub indexes real-time event sources.

Where Real-Time Hub Fits¶

flowchart LR
    subgraph Sources["📡 Event Producers"]
        IOT["IoT Devices"]
        DB["Database CDC"]
        APP["Applications"]
        AZ["Azure Events"]
    end

    subgraph Hub["🏠 Real-Time Hub"]
        CAT["Event Catalog"]
        TOP["Managed Topics"]
        DER["Derived Streams"]
    end

    subgraph Consumers["🎯 Event Consumers"]
        ES["Eventstreams"]
        EH["Eventhouse (KQL)"]
        DA["Data Activator"]
        LH["Lakehouse"]
    end

    Sources --> Hub --> Consumers

    style Hub fill:#8E44AD,stroke:#6C3483,color:#fff

Key Capabilities¶

🏗️ Architecture¶

Component Architecture¶

flowchart TB
    subgraph RTHub["📡 Real-Time Hub"]
        subgraph Sources["Event Sources"]
            DS["Data Streams<br/>(Eventstream outputs)"]
            MS["Microsoft Sources<br/>(Event Hubs, IoT Hub, CDC)"]
            FE["Fabric Events<br/>(job runs, refreshes)"]
        end

        subgraph Catalog["Event Catalog"]
            DISC["Discovery &<br/>Search"]
            PREV["Data Preview"]
            META["Metadata &<br/>Schema"]
        end

        subgraph Actions["Consumer Actions"]
            ES["Create Eventstream"]
            EH["Send to Eventhouse"]
            DA["Set Alert"]
            LH["Send to Lakehouse"]
        end
    end

    Sources --> Catalog --> Actions

    style RTHub fill:#F5EEF8,stroke:#8E44AD
    style Catalog fill:#D2B4DE,stroke:#8E44AD

Three Event Categories¶

⚙️ Event Sources¶

Data Streams (Eventstream Outputs)¶

Any Eventstream output is automatically available in the Real-Time Hub:

Eventstream: casino-slot-events
  Input: Azure Event Hub (casino-events-hub)
  Output 1: Eventhouse (slot_analytics)     → appears in Real-Time Hub
  Output 2: Lakehouse (lh_bronze)           → appears in Real-Time Hub

Microsoft Sources¶

Connect directly from the Hub:

Fabric Events¶

Monitor Fabric platform activity:

Connecting a Source¶

1. Open Real-Time Hub from the left nav
2. Click + Get events
3. Select source type (Event Hub, IoT Hub, etc.)
4. Configure connection:

   Source: Azure Event Hub
   Event Hub namespace: casino-events.servicebus.windows.net
   Event Hub name: slot-events
   Consumer group: $Default
   Authentication: Shared Access Key
   

5. Preview data to verify schema
6. Choose destination: Create Eventstream, Send to Eventhouse, or Set Alert

🔄 Streams and Topics¶

Derived Streams¶

Create filtered views of existing streams for specific consumers:

Parent stream: casino-slot-events (all machines, all casinos)

Derived stream 1: casino-a-high-value
  Filter: casino_id = "CASINO-A" AND coin_in > 1000

Derived stream 2: jackpot-events
  Filter: event_type = "JACKPOT"

Derived stream 3: compliance-monitoring
  Filter: amount >= 3000

Stream Sharing¶

Share streams across workspaces:

Real-Time Hub → Stream → Share
  Share with:
    ☑ casino-floor-ops-workspace (Read)
    ☑ casino-compliance-workspace (Read)
    ☑ casino-analytics-workspace (Read)

🎰 Casino Implementation¶

Casino Floor Event Topology¶

flowchart TB
    subgraph Floor["🏢 Casino Floor"]
        SM["Slot Machines"]
        TG["Table Games"]
        SEC["Security Cameras"]
        POS["POS Systems"]
    end

    subgraph Hub["📡 Real-Time Hub"]
        S1["slot-events stream"]
        S2["table-game-events stream"]
        S3["security-events stream"]
        S4["pos-events stream"]
        D1["high-value-slots<br/>(derived)"]
        D2["jackpot-events<br/>(derived)"]
        D3["compliance-events<br/>(derived)"]
    end

    subgraph Consumers["🎯 Consumers"]
        OPS["Floor Ops<br/>Dashboard"]
        COMP["Compliance<br/>Monitoring"]
        AI["AI/ML<br/>Models"]
        STORE["Lakehouse<br/>Storage"]
    end

    SM --> S1
    TG --> S2
    SEC --> S3
    POS --> S4
    S1 --> D1 --> OPS
    S1 --> D2 --> OPS
    S1 --> D3 --> COMP
    S1 --> AI
    S1 --> STORE
    S2 --> STORE

Available Streams in Hub¶

🏛️ Federal Implementation¶

Multi-Agency Event Hub¶

Cross-Agency Data Sharing¶

Real-Time Hub enables agencies to share event streams:

NOAA severe-weather-alerts → shared with:
  - DOT/FAA (flight impact assessment)
  - USDA (crop impact monitoring)
  - EPA (air quality correlation)
  - DOI (natural disaster response)

⚠️ Limitations¶

📚 References¶

📊 Capacity Events in Real-Time Hub (Preview)¶

Announced at FabCon Atlanta March 2026, Capacity Events bring real-time visibility into Fabric capacity usage directly into the Real-Time Hub. This enables proactive workload management by surfacing:

• Throttling Events: Alerts when capacity throttling begins or ends
• Usage Spikes: Real-time CU consumption exceeding configurable thresholds
• Performance Degradation: Latency increases across workloads detected automatically
• Capacity Utilization: Continuous percentage utilization stream updated every 30 seconds

How It Works¶

Capacity events appear as a new system event source in the Real-Time Hub. Once enabled:

1. Navigate to Real-Time Hub → System Events → Capacity Events
2. Select the target Fabric capacity
3. Configure event filters (throttling, utilization thresholds, latency)
4. Route events to an Eventstream, KQL Database, or Data Activator reflex
5. Build dashboards or automated responses on the event stream

Casino Use Case¶

Casino gaming floors generate unpredictable compute spikes during peak hours (Friday/Saturday nights, major sporting events). Capacity events enable:

• Real-time alerts when slot telemetry ingestion causes throttling
• Automated scaling triggers via Data Activator (e.g., pause low-priority refreshes)
• Historical analysis of capacity patterns for F64 right-sizing decisions
• Correlation of CU spikes with specific workloads (e.g., CTR batch processing)

Federal Use Case¶

Federal agencies with shared Fabric tenants (e.g., USDA + NOAA sharing capacity) can monitor per-agency consumption and prevent one agency's batch jobs from throttling another's real-time dashboards. Capacity events provide the telemetry foundation for chargeback models and fair-use enforcement across organizational boundaries.

🔗 Related Documents¶

• Real-Time Intelligence — Eventstreams and Eventhouse for analytics
• Data Activator — Alerting triggered from Real-Time Hub events
• Data Sharing & Federation — Cross-workspace sharing
• Architecture — System architecture

📝 Document Metadata - Author: Documentation Team - Reviewers: Real-Time Intelligence, Platform - Classification: Internal - Next Review: 2026-07-21