Azure Messaging - Standards Matter

In this session, Microsoft's lead architect for Messaging, Clemens Vasters, will provide an overview of the Azure Messaging services (Event Hubs, Event Grid, and Service Bus) in the industry context and with a lens on interoperability. You will learn about the various standardization activities that Microsoft is involved in to help drive interoperability across different platform and cloud vendors, and how those standards surface in Microsoft's platform services. You'll be getting a little crash course in some key protocols like HTTP, AMQP, MQTT, Kafka and will learn how they overlap or complement each other. You'll also learn about higher level standards such as CNCF CloudEvents and OPC UA that provide a layer of abstraction for special use-case scenarios.

Type-safe, polyglot messaging and eventing: Stream/Queue/Topic "of T"

A long time ago, the introduction of templates and generics helped to make applications written in languages like C++, C#, Java far more robust by allowing for common data structures like collections to handle specific types. More recent designs like Rust and TypeScript started with support for generics at inception. Especially in business applications, strong type systems help avoiding simple errors like omitting fields or losing information due to mismatched data types.

In asynchronous, inter-application messaging and eventing, we don't yet have equivalently strong type-safety model that works end-to-end and across multiple vendor platforms, but the building blocks are in the works.

In this session, Clemens Vasters, Product and Standards Architect in the Microsoft Messaging and Real-Time Analytics team and member of the CNCF CloudEvents and xRegistry projects, discusses the schema models, schema registries, metadata projects, code generators, and other tooling, and an interoperability roadmap towards "Stream of T"

45+ minutes required

Azure Messaging - Streams, Queues, and Pub Sub

Azure Messaging services, including Azure Service Bus, Azure Event Grid, Azure Event Hubs, Azure Relay, Azure Stream Analytics, and Fabric Event Streams, offer a suite of tools for reliable message delivery and event-driven architecture. This session will provide a clear understanding of each service’s capabilities and how they can be integrated to build scalable, resilient applications.

We’ll discuss the selection process for the right messaging tool based on your application’s needs, supported by real-world examples. The focus will be on practical guidance and best practices for implementing a cohesive messaging strategy on Azure, and we will look at the latest news from each product.

Join us to learn how Azure’s messaging services can streamline your application’s communication layer, fostering efficiency and innovation in your cloud-based solutions.

Bringing Clarity to Event Streams: Enabling Analytics and AI Through Rich Metadata

Event streams are central to delivering actionable insights for analytics and AI. However, without rich metadata to contextualize these streams, data consumers often struggle with inconsistencies, poor quality, and a lack of interpretability. By leveraging metadata as a cornerstone of event stream architecture, organizations can unlock clarity in their streaming data, bridging the gap between raw event streams and actionable insights. This clarity not only enhances the quality and usability of event streams but also ensures alignment between developers, analysts, and AI systems, enabling them to extract maximum value from data in motion.

This session explores how rich metadata transforms event streams into a reliable foundation for advanced analytics and AI workflows. We will examine the role of schema definitions, semantic annotations, and cataloging practices in improving data quality and context. Attendees will gain practical insights into building metadata-enriched event stream pipelines, fostering collaboration between data producers and consumers, and enabling AI models to operate with greater precision. Whether you are developing real-time applications or enhancing decision-making with AI, this talk will provide actionable strategies to harness the power of rich metadata in event streaming.

Generating code and projects in all sorts of languages with Python and Jinja2

This session summarizes about two years of intense work building code generators based on data schemas and, indeed, generators for complete projects, using Python and the Jinja2 templating framework with C++, C#, Java, JavaScript, TypeScript, Rust, Go, and (of course) Python as target languages. You'll learn about the considerations for the code generation input metadata, the required pre- and post-processing and a series of Jinja2 extensions that specifically help with code-generation.

The session is 95% "walking through the code" and expects familiarity with Python and a keen interest in code generation.

Streamifying Reference Data for Temporal Consistency with Telemetry Events

Reference data provides critical context for interpreting telemetry, yet its traditional static handling limits temporal consistency in dynamic systems. Unlike real-time telemetry, reference data is often updated asynchronously in bulk, making it difficult to align with rapidly changing events. For example, transit systems rely on GTFS files for schedules and GTFS-RT for real-time updates, creating mismatched formats and update frequencies. To address this, reference data must be "streamified," where every state change is treated as an event. By detecting and signaling changes—such as through hashing techniques—reference data can become temporally consistent and integrate seamlessly with telemetry for point-in-time accuracy.

Reframing reference data as event streams enables a unified approach to real-time processing. Standardizing reference data events with CNCF CloudEvents embeds critical metadata for temporal alignment, ensuring consistent context for both real-time and historical analysis. This transformation unlocks new potential for industries like IoT, transit, and healthcare, where accurate, time-aligned reference data is essential. By treating reference data as a first-class citizen in event-driven systems, organizations can achieve higher precision and reliability in their analytics.

Real-time data journeys. Kafka, MQTT, AMQP, HTTP, Web Sockets; protocols, my, oh, my!

Before you can work with your real-time data in Microsoft Fabric, the data (obviously) first needs to get there. The journey of telemetry and streaming data from its origin can be facilitated by a choice of application protocols, each tailored to specific use cases and challenges. This talk explores the landscape of real-time communication with a deep dive into Kafka, MQTT, AMQP, HTTP, and Web Sockets. We’ll examine the strengths and trade-offs of each protocol, focusing on their roles in streaming, messaging, telemetry, and bidirectional communication. Through practical examples and architecture patterns, you’ll learn how to choose the right protocol for your application and how you ultimately land the data into Microsoft Fabric. Join us as we demystify the protocol maze and navigate the complexities of real-time data exchange.

From Observations to Actions: Real-Time Intelligence with Microsoft Fabric

Microsoft Fabric Real Time Intelligence is a transformative technology designed to empower businesses with real-time data insights and analytics. This session will cover the journey of Real Time Intelligence from its inception to its current state, highlighting key features, capabilities, and the impact it has on operational efficiency and decision-making processes. We will also highlight customer success stories and showcase how you can revolutionize your data strategy and unlock new opportunities for growth.

Endpoints, Messages, Schemas: CNCF xRegistry

The CNCF xRegistry project is an offspring of the graduated CNCF CloudEvents project, motivated by the need to formally declare which events can be raised by services and which are available to handle. This session provides an overview of the xRegistry metadata model, its API and the mirroring document format, dives into reference implementations, explains its use in products, and shows you how you can leverage xRegistry to build robust and type-safe event pipelines.

What is a Message Queue and when and why would I use it?

I'm the lead architect in central engineering of Microsoft Azure's messaging and eventing services fleet and one observation we keep making is that developers don't know about queues and what to do with them. This session is a 101 basic introduction into what a message queue is, what solution patterns it helps with, and what to do with features like dead-lettering and expiry. This session isn't about what's new, it's about the whole concept. The talk is not product specific and applies to anyone's product's and services, even though I'll certainly mention our own.

Target audience is all developers who haven't been using queues yet or are mistaking some other technology (Kafka?) for a message queue. Min duration is 40 minutes.

Azure Service Bus - Behind the scenes

Azure Service Bus is the planet's most powerful hyperscale PaaS message queuing platform. Running in "ring zero" of the Azure platform, Azure Service Bus is the reliability bedrock for many Azure services, but also for Microsoft's own internal commercial systems, Microsoft 365, Microsoft Dynamics, and of course tens of thousands of customer solutions. Your airport check-in, settling the bill for your online purchase, the package delivery at your doorstep, and unlocking your vehicle with a mobile app may all run through Azure Service Bus. In this session we will pull away the curtain and give you a deep-dive look at how Service Bus is built and run.

Azure Event Hubs - Behind the Scenes

Azure Event Hubs is a hyperscale PaaS event stream broker with protocol support for HTTP, AMQP, and Apache Kafka RPC that accepts and forwards several trillion (!) events per day and is available in all global Azure regions. This session is a look behind the curtain where we dive deep into the architecture of Event Hubs and look at the Event Hubs cluster model, resource isolation, and storage strategies and also review some performance figures.