Streaming Design Notes in Fabric: keeping streaming dashboards useful after week one
I focused on making delivery decisions auditable and repeatable—documenting intent, success criteria, and rollback paths to reduce tribal knowledge.
41 articles
I focused on making delivery decisions auditable and repeatable—documenting intent, success criteria, and rollback paths to reduce tribal knowledge.
I worked on smoothing the handoff between data engineering and AI teams—standardizing feature contracts, embedding validation, and adding lightweight…
I focused on making delivery decisions auditable and repeatable—documenting intent, success criteria, and rollback paths to reduce tribal knowledge.
I spent the day reducing cognitive overhead for engineers and analysts—introducing clearer table contracts, simpler failure modes, and concise runbooks that…
I turned implicit processes into explicit operating rules—defining owners, acceptance tests, and lightweight runbooks so teams can move confidently and…
I spent the day reducing cognitive overhead for engineers and analysts—introducing clearer table contracts, simpler failure modes, and concise runbooks that…
Here's what it is, how it works, and a pattern that's running in production. Traditional analytics is batch-oriented. Data lands in storage, gets processed…
Real-time dashboards are now straightforward with Azure SignalR. The serverless model means you only pay for messages, making it cost-effective for most…
Data Activator monitors data streams and triggers actions based on conditions you define. It connects to Power BI reports, Eventstreams, and Fabric data…
Streaming inference brings AI insights to real-time data. Design for throughput, latency, and reliability from the start.
Real-time AI requires discipline around latency. Design for the worst case and optimize for the common case.
Real-time AI requires careful architecture to balance latency, cost, and quality. Start with simple use cases and optimize based on actual performance data.
Eventstream enhancements make real-time analytics more powerful and accessible. Start with simple streaming scenarios and progressively add complexity.
CDC captures row level changes (inserts, updates, deletes) from source systems: The simplest option for supported sources: For more control over the CDC…
For sub second latency: For minute level latency with relational sources: Combine streaming and batch for different data types: Distribute data to multiple…
Mirroring creates a continuously synchronized copy of your operational database in OneLake as Delta Lake tables: As of GA: Azure SQL Database : Full support…
Latency has multiple components: 1. Network latency : Request travel time 2. Queue time : Waiting for processing 3. Time to first token (TTFT) : Initial…
Data Activator monitors your data and triggers actions when conditions are met: A Reflex is a Data Activator item containing: Objects : Entities to monitor…
Online feature serving enables real-time ML inference by providing low-latency access to features. This guide covers setting up and using Databricks online…
Build real-time transcription applications using Azure AI Speech services with low latency and high accuracy.
Tomorrow we'll explore Real Time Dashboards for visualizing streaming data. Eventstream Documentation Event Hub Integration Streaming Best Practices
Streaming improves the user experience significantly. Tomorrow, I will cover error handling for AI applications.
Mirroring simplifies real-time data replication to Fabric. Tomorrow, I will cover Git integration in Fabric.
Combining Citus's distributed query power with rollup tables and caching enables dashboards that handle millions of events per second.
The new visual editor allows building streaming pipelines without SQL: 1. Drag and drop input/output connections 2. Visual transformations and aggregations…
Azure's real time analytics stack enables instant insights from millions of events per second, powering live dashboards and alerts.
Azure Web PubSub offers native WebSocket support with powerful server side messaging APIs, ideal for IoT, gaming, and collaboration apps.
Azure SignalR Service scales real time applications effortlessly, handling millions of connections with automatic load balancing.
Long polling is the workaround for near-real-time notifications in environments where WebSocket connections aren't available—the client sends an HTTP…
Feature SSE WebSocket Direction Server to client Bidirectional Protocol HTTP WebSocket Reconnection Automatic Manual Binary data No Yes SSE offers a simple,…
GraphQL subscriptions provide elegant real time communication, ideal for dashboards, notifications, and live updates.
Voice and video calling Chat messaging SMS messaging Email communication Phone number management Create and configure your Communication Services resource:…
DLT streaming tables use append mode by default. For aggregations: DLT handles checkpointing automatically: Streaming tables in Delta Live Tables provide:…
All three happen with minimal latency for real-time conversations.
Hybrid tables provide the perfect balance between real-time data freshness and historical data performance.
Azure Stream Analytics sits in a specific niche that I've come to appreciate: SQL-native stream processing for teams who shouldn't have to learn Spark or…
Azure Web PubSub is the lower-level real-time service that sits alongside SignalR. Where SignalR abstracts the transport and provides a hub model for your…
I've tried to roll my own WebSocket fanout exactly once. It worked beautifully for 200 concurrent connections and melted at 2,000. Azure SignalR Service…
Azure Communication Services launched late last year, and "build your own video calling app on the same backbone as Teams" is the pitch that makes it click…
Stream Analytics is one of those services I keep recommending and clients keep being surprised by — SQL on top of an event firehose, with windowing and…