Real-Time Analytics on Azure

Real-time services on Azure

Simple explanation

Real-time analytics is like a control room — screens showing live data, alarms going off when something’s wrong.

Azure provides three key services for this: Event Hubs catches the incoming data (like a massive funnel), Stream Analytics processes it in motion (filtering, aggregating, alerting), and Fabric Real-Time Intelligence does the same inside the Fabric platform.

The real-time pipeline

Data Sources → Event Ingestion → Stream Processing → Output
(IoT, apps)    (Event Hubs)      (Stream Analytics)   (Dashboard, database, alert)

Azure Event Hubs

The front door for streaming data. Event Hubs is a big data streaming platform that can ingest millions of events per second.

Key features:

Capture millions of events per second
Partitioned for parallel processing
Retain events for replay (1-7 days standard, up to 90 days with Premium/Dedicated)
Works with Kafka protocol (existing Kafka apps can connect)

Tom’s example: GPS updates from 200 trucks flow into Event Hubs → 200 events every 10 seconds = 1,200 events/minute.

Azure Stream Analytics

A real-time query engine that processes streaming data using SQL-like queries. It reads from Event Hubs (or IoT Hub), applies transformations, and writes results to outputs.

-- Alert when a delivery is more than 30 minutes late
SELECT DriverID, DeliveryID, Delay
FROM GPSStream
WHERE Delay > 30

Key features:

SQL-like query language (easy for analysts)
Built-in temporal functions (windows, time-based aggregation)
Multiple outputs (Power BI, Blob Storage, Cosmos DB, SQL Database)
No infrastructure to manage (serverless)

Fabric Real-Time Intelligence

Microsoft Fabric’s built-in real-time analytics capability. It includes:

Eventstreams — capture and route streaming data within Fabric
KQL Database — a database optimised for real-time queries using Kusto Query Language
Real-Time dashboards — live visualisations connected to streaming data

When to use Fabric Real-Time vs Stream Analytics:

If you’re already in Fabric → use Fabric Real-Time Intelligence
If you need a standalone streaming service → use Stream Analytics

Azure real-time analytics services
Feature	Event Hubs	Stream Analytics	Fabric Real-Time Intelligence
Role	Ingest streaming data	Process streaming data	End-to-end real-time analytics in Fabric
What it does	Catches millions of events/sec	SQL-like queries on data in motion	Ingest, query, and visualise in one platform
Query language	N/A (ingestion only)	SQL-like (ASA SQL)	KQL (Kusto Query Language)
Best for	High-volume event capture	Real-time filtering and alerting	Real-time analytics within Fabric

Azure IoT Hub vs Event Hubs

Both ingest streaming data, but for different scenarios:

Event Hubs: General-purpose event streaming (application logs, clickstreams, telemetry)
IoT Hub: Purpose-built for IoT devices — adds device management, per-device authentication, and cloud-to-device messaging

For DP-900, know that IoT Hub exists for IoT scenarios and Event Hubs is the general-purpose option.

Exam tip: real-time service selection

“Ingest millions of events per second” → Event Hubs
“Process streaming data with SQL-like queries” → Stream Analytics
“Real-time analytics inside Microsoft Fabric” → Fabric Real-Time Intelligence
“IoT device management + telemetry ingestion” → IoT Hub
“Real-time Power BI dashboard” → Stream Analytics output to Power BI, or Fabric real-time dashboard

Flashcards

Question

What is Azure Event Hubs?

Click or press Enter to reveal answer

Answer

A big data streaming platform for ingesting millions of events per second. It's the front door for streaming data — capturing events from applications, IoT devices, and services before they're processed.

Click to flip back

Question

What is Azure Stream Analytics?

Click or press Enter to reveal answer

Answer

A real-time query engine that processes streaming data using SQL-like queries. It reads from Event Hubs or IoT Hub, applies transformations and filters, and sends results to outputs like Power BI or databases.

Click to flip back

Question

What is Fabric Real-Time Intelligence?

Click or press Enter to reveal answer

Answer

Microsoft Fabric's built-in real-time analytics service. It includes Eventstreams (data capture), KQL Database (real-time queries), and real-time dashboards — all within the Fabric platform.

Click to flip back

Knowledge check

Knowledge Check

Pacific Freight needs to process GPS events from 200 trucks in real time and alert dispatchers when deliveries are more than 30 minutes late. Which Azure service processes the streaming data?

Knowledge Check

A smart factory generates 5 million sensor readings per second. Which service should capture this high-volume data stream before processing?

Next up: Power BI: See Your Data — the tool that turns all this data into visual insights.