Bad Data Architectures: Lessons Learned the Hard Way en

Every bad data architecture starts with a good intention.

A quick fix becomes permanent. A temporary integration survives for years. A "simple" reporting solution grows into a business-critical dependency nobody dares to touch. Before long, teams are drowning in complexity, performance problems, duplicate data, unclear ownership, and endless technical debt.

In this session, we'll dissect real-world architecture anti-patterns from databases that became integration hubs and data lakes that turned into swamps, to over-engineered real-time platforms and AI projects built on shaky foundations. Through stories, diagrams, and painful lessons learned, you'll discover why these architectures fail, how the warning signs can be spotted early, and what successful teams do differently.

Whether you're a DBA, data engineer, architect, or developer, you'll leave with practical principles for designing data platforms that remain scalable, maintainable, and resilient long after the first project goes live.

Data API Builder in the Real World: From Database to API in 60 Minutes en

Modern applications, integration platforms, and AI solutions all need access to data, yet many teams still spend weeks building and maintaining custom APIs. Data API Builder (DAB) offers a different approach by enabling developers to expose SQL Server and Azure SQL data through secure REST and GraphQL endpoints with minimal effort.

In this demo-driven session, we'll start with a database and build a production-ready API layer from scratch. Along the way, you'll learn how to define entities, configure authentication and authorization, expose relationships, control CRUD operations, and secure access using role-based permissions. We'll also explore deployment options, integration scenarios, and common pitfalls encountered in real-world projects.

Whether you're a developer, data engineer, architect, or DBA, you'll leave with practical guidance, proven patterns, and a clear understanding of when Data API Builder can accelerate delivery and when a traditional custom API remains the better choice.

10 Data Architecture Mistakes You'll Make Unless You Learn from Mine en

Microsoft Fabric makes it easier than ever to get data into the hands of users. It also makes it surprisingly easy to create workspace sprawl, inconsistent security models, duplicated semantic models, unclear ownership, and governance headaches that only become visible once hundreds of users depend on the platform.

In this session, we'll examine real-world governance mistakes that organizations commonly make when adopting Microsoft Fabric. Through practical scenarios, architecture diagrams, and lessons learned, you'll discover how seemingly harmless decisions in the early stages of adoption can become major operational, security, and compliance challenges later.

Whether you're a Data Engineer, Architect, Platform Owner, or Fabric Administrator, you'll leave with actionable guidance for building a Fabric environment that scales without becoming a data swamp.

Warum „SQL Server oder Microsoft Fabric?“ die falsche Frage ist de en

SQL Server oder Microsoft Fabric? Das ist die falsche Frage. Erfolgreiche Datenplattformen setzen heute nicht auf eine einzelne Technologie, sondern auf die richtige Kombination aus beiden. In dieser Session betrachten wir praxisnahe Architekturmuster, die SQL Server, Azure SQL und Microsoft Fabric für Analytics, KI und Governance miteinander verbinden. Anhand realer Szenarien zeige ich, wo die jeweiligen Stärken liegen, wie die Plattformen zusammenspielen und wie eine zukunftssichere Modernisierungsstrategie ohne vollständige Migration aussehen kann.

Orchestrate Like a Pro: Dynamic Pipelines in ADF en de

Think your data pipelines are smart? Think again! In this session, Azure Data Factory gets a serious upgrade. Participants will see how to orchestrate complex data loads using just a handful of pipelines and a single JSON file, making workflows more flexible, reusable, and easier to manage.

About 95% of the session is live demos, walking through real-world scenarios. Participants will learn how to use parameters and dynamic content to reduce the number of pipelines needed, how to make pipelines adapt to different sources and destinations automatically, and how to structure pipelines and JSON files for maximum reusability. Practical examples will show how to achieve faster, cleaner, and more maintainable operations.

Whether participants know the basics of ADF or just want to see magic in action, this session provides practical techniques and inspiration to transform the way pipelines are built and deployed. Attendees will leave with actionable insights they can immediately apply to their own workflows, making data loading smarter and surprisingly fun.

Smooth Deployments Ahead - CI/CD with Azure Data Factory and DevOps en de

Ready to take Azure Data Factory (ADF) pipelines to the next level? This session shows how to supercharge them with continuous integration and continuous deployment (CI/CD) in Azure DevOps. Participants will learn how to structure ADF repositories smartly and pick up practical strategies to manage and deploy pipelines without the usual headaches.

The session walks through building and releasing pipelines across dedicated Development and Production environments, ensuring workflows run smoothly and productivity gets a serious boost.

Note: the focus is on the operational side of CI/CD, not on building ADF pipelines from scratch. A basic knowledge of ADF is recommended. This session makes deployment simpler, faster, and gives data pipelines the DevOps treatment they deserve—turning a complex process into a practical, manageable, and surprisingly enjoyable experience.

Turbocharging Data Management: Mastering Performance with Massive Tables en

Navigating the complexities of extremely large SQL tables can be daunting, especially when performance issues arise. In this demo-driven session, we’ll dive into a compelling business case involving the processing of an astounding 60 million new data records per day on a relational on-premise SQL Server database.

We will kick off with an exploration of essential concepts such as partitions and indices. Whether you’re a seasoned user or new to these techniques, this segment will provide valuable insights and practical strategies for optimizing your database performance.

After that, witness the power of real-time data management in action. We’ll demonstrate how to efficiently archive millions of records without the downtime typically associated with delete operations. You'll see firsthand how to seamlessly move data across various storage locations instantly, enhancing your system’s agility and responsiveness.

Join us to learn how to unlock the full potential of your large data sets and transform your approach to data management!

Working Smarter: Streamlined automation and development en

In our professional lives, we often face repetitive tasks and time-consuming processes that distract us from what truly matters. In this session, I will demonstrate how tools like Elgato Stream Deck and Touch Portal can be used to implement simple automations that significantly enhance workflow efficiency—regardless of your specific field of work.

The session will include SQL-specific examples relevant to developers and solution architects, as well as general use cases like optimizing collaboration and productivity with tools such as Microsoft Teams. This session is for anyone looking to optimize their daily routines through smart automation solutions, whether in IT, management, or other areas.

Takeaways:
- Efficient automation for various work environments
- SQL-specific use cases to streamline the developer's daily tasks
- Flexible application of automation tools for general tasks like Microsoft Teams

This session is ideal for anyone seeking to enhance their workflow with creative automation solutions for their day-to-day tasks, with a special focus on SQL development.

While the featured tools are not free, this is NOT a commercial session. The focus is entirely on sharing experiences and practical tips to improve everyday work life.

Data on Tap: Streamlining Processing with Azure Service Bus and Functions en

In a world where data streams are constantly flowing, timely processing becomes crucial. This session explores the efficient handling of real-time data using Azure Service Bus and Azure Functions, with a strong emphasis on minimizing processing lag.

Through a mostly demo-based approach, I’ll showcase how to build a responsive data pipeline capable of ingesting, processing, and directing data into an SQL database with minimal delay. Using simulated sales data from Munich’s world-famous beer festival, you’ll experience firsthand the challenges and solutions in real-time data processing. From message handling in Azure Service Bus to using Azure Functions to manage incoming data, this session provides practical insights into creating a scalable and efficient workflow.

Ideal for developers and data engineers, you’ll leave with actionable techniques to streamline your own real-time data processing. Join us to unlock the full potential of real-time data in Azure!

Window Functions - The Superpowers of SQL en de

SQL has plenty of tools, but window functions are the one most people never quite get around to learning, until they're stuck three subqueries deep and wondering why this is so hard. Turns out it doesn't have to be.

For this session we're using an Oktoberfest dataset, because honestly, beer tents and crowds make for a pretty fun way to look at data. We'll poke around questions like which tent draws the biggest crowd, how drinking picks up (or slows down) over the course of the evening, and which tent ends up on top by the end of the night. Window functions make all of this pretty painless, no piling up subqueries or joins to get there.

We'll go through it together at a relaxed pace: running totals, moving averages, comparing a row to the ones around it. Nothing rushed, just working through real examples using the Oktoberfest data so it's easier to remember than the usual table A and table B.

Come as you are, whether you've been writing SQL for years or you're still getting comfortable with it. By the end you'll probably just wonder why you didn't pick this up sooner.

Partitioning in SQL Server - Do not Panic, It's Just Slices en de

Partitioning often sounds like a topic reserved for SQL wizards managing massive tables, but it does not have to be intimidating. In this session, participants will discover how partitioning can transform everyday SQL work, making queries faster, tables easier to manage, and even complex reporting tasks simpler.
The session starts from the basics: what a partition is, how to set it up, and why it matters. Hands-on examples show practical benefits that anyone using SQL Server can apply immediately, even without terabytes of data. By the end, participants will see partitioning not as an advanced trick, but as a powerful tool that makes SQL work smarter, not harder.
This session is perfect for those who have been avoiding partitioning, offering a clear, approachable, and surprisingly fun introduction that turns a seemingly complex topic into something useful and immediately applicable.

Window Functions - Die Superkräfte von SQL en de

SQL hat jede Menge Werkzeuge, aber Window Functions sind die, die die meisten nie so richtig lernen, bis sie irgendwann drei Subqueries tief feststecken und sich fragen, warum das alles so kompliziert sein muss. Muss es aber gar nicht.

Für diese Session nehmen wir einen Oktoberfest-Datensatz, denn ehrlich gesagt sind Bierzelte und Menschenmassen eine ziemlich unterhaltsame Art, sich mit Daten zu beschäftigen. Wir schauen uns Fragen an wie: Welches Zelt zieht die meisten Besucher an? Wie verändert sich der Bierkonsum im Laufe des Abends? Und welches Zelt liegt am Ende vorne? Mit Window Functions lässt sich das alles ziemlich entspannt beantworten, ganz ohne einen Berg an Subqueries oder Joins.

Wir arbeiten uns in Ruhe gemeinsam durch: laufende Summen, gleitende Durchschnitte, der Vergleich einer Zeile mit ihren Nachbarn. Nichts wird gehetzt, wir gehen einfach anhand echter Beispiele durch, und mit den Oktoberfest-Daten bleibt das Ganze leichter im Kopf als mit den üblichen Tabelle A und Tabelle B.

Am Ende wirst du dich wahrscheinlich fragen, warum du dir das nicht schon viel früher angeschaut hast.

Partitionierung in SQL Server - Keine Panik, nur Scheiben en de

Partitionierung klingt oft wie ein Thema für SQL-Zauberer, die riesige Tabellen verwalten, muss aber keinesfalls einschüchternd sein. In dieser Session entdecken die Teilnehmer, wie Partitionierung den Alltag mit SQL verändern kann: Abfragen werden schneller, Tabellen lassen sich leichter verwalten und sogar komplexe Reporting-Aufgaben werden einfacher.

Die Session beginnt bei den Grundlagen: Was ist eine Partition, wie richtet man sie ein und warum ist sie wichtig? Praxisnahe Beispiele zeigen Vorteile, die jeder SQL-Server-Nutzer sofort anwenden kann, selbst ohne Terabytes an Daten. Am Ende werden die Teilnehmer Partitionierung nicht mehr als fortgeschrittenen Trick, sondern als mächtiges Werkzeug sehen, das SQL-Arbeit smarter, nicht härter macht.

Diese Session ist ideal für alle, die Partitionierung bisher gemieden haben, und bietet eine klare, zugängliche und überraschend unterhaltsame Einführung in ein scheinbar komplexes Thema, das sofort praktisch anwendbar ist.

Azure Data Factory Pipelines - Mit CI/CD stressfrei deployen en de

Bereit, Azure Data Factory (ADF)-Pipelines auf das nächste Level zu bringen? Diese Session zeigt, wie man sie mit Continuous Integration und Continuous Deployment (CI/CD) in Azure DevOps richtig aufpeppt. Die Teilnehmer lernen, wie man ADF-Repositories clever strukturiert und erhalten praktische Strategien, um Pipelines zu verwalten und bereitzustellen, ganz ohne die üblichen Kopfschmerzen.

Die Session führt Schritt für Schritt durch den Aufbau und die Bereitstellung von Pipelines in getrennten Entwicklungs- und Produktionsumgebungen, sodass Workflows reibungslos laufen und die Produktivität spürbar steigt.

Hinweis: Der Fokus liegt auf dem operativen Teil von CI/CD, nicht auf dem Erstellen von ADF-Pipelines von Grund auf. Grundkenntnisse in ADF werden empfohlen. Diese Session macht das Deployment einfacher, schneller und verleiht den Datenpipelines das DevOps-Treatment, das sie verdienen – ein komplexer Prozess wird praktisch, handhabbar und überraschend unterhaltsam.

Mastering JSON in SQL Server: From Complexity to Everyday Power en de

JSON in SQL Server often sounds complicated, but it is actually flexible, practical, and surprisingly fun once participants get hands-on. This topic is increasingly important because modern applications generate a lot of semi-structured data, and knowing how to handle it efficiently in SQL Server provides a major advantage in real-world projects.
The session guides participants step by step through storing JSON in SQL Server, querying it, and combining it with existing tables, without drowning in theory slides. Starting with the basics, participants will learn how to store JSON properly, access individual values or entire arrays, and join JSON data with relational tables to uncover real insights. Mini hands-on examples, such as analyzing log data or importing orders, make the concepts immediately tangible.
By the end of the session, participants will not only understand JSON, but also be able to use it to save time and make SQL queries more flexible, turning a seemingly complex topic into a practical, everyday tool.

JSON in SQL Server - Daten bauen wie mit LEGO en de

JSON in SQL Server klingt oft kompliziert, ist aber tatsächlich wie LEGO für Daten: flexibel, praktisch und überraschend spaßig, sobald man selbst Hand anlegt. Dieses Thema gewinnt zunehmend an Bedeutung, da moderne Anwendungen viele semi-strukturierte Daten erzeugen. Wer weiß, wie man diese effizient in SQL Server verarbeitet, hat einen klaren Vorteil in realen Projekten.

In dieser Session werden die Teilnehmer Schritt für Schritt angeleitet, wie man JSON in SQL Server speichert, abfragt und mit bestehenden Tabellen kombiniert, ohne sich in Theorie zu verlieren. Beginnend bei den Grundlagen lernen die Teilnehmer, wie man JSON korrekt speichert, auf einzelne Werte oder ganze Arrays zugreift und JSON-Daten mit relationalen Tabellen verbindet, um echte Insights zu gewinnen. Kleine praktische Beispiele, wie das Analysieren von Logdaten oder das Importieren von Bestellungen, machen die Konzepte sofort greifbar.

Am Ende der Session werden die Teilnehmer JSON nicht nur verstehen, sondern es auch nutzen können, um Zeit zu sparen und SQL-Abfragen flexibler zu gestalten. Ein scheinbar komplexes Thema wird so zu einem praktischen Werkzeug für den Alltag.

Wie ein Profi orchestrieren: Dynamische Pipelines in ADF en de

Denkst du, deine Datenpipelines sind schon clever? In dieser Session zeigt Azure Data Factory, wie viel mehr möglich ist. Mit nur wenigen Pipelines und einer einzigen JSON-Datei lernen die Teilnehmer, wie sich komplexe Datenladeprozesse deutlich flexibler, wiederverwendbarer und leichter zu verwalten gestalten lassen.

Fast die gesamte Session besteht aus Live-Demonstrationen, die durch praxisnahe Szenarien führen. Dabei erfahren die Teilnehmer, wie Parameter und dynamische Inhalte die Anzahl der benötigten Pipelines reduzieren, wie Pipelines automatisch an unterschiedliche Quellen und Ziele angepasst werden und wie sich Pipelines und JSON-Dateien optimal strukturieren lassen, um sie wiederverwendbar zu machen. Die Beispiele zeigen auf anschauliche Weise, wie sich Prozesse schneller, sauberer und einfacher warten lassen.

Ob mit Grundkenntnissen in Azure Data Factory oder als neugieriger Zuschauer: Die Teilnehmer erhalten praktische Tipps und Inspiration, um den Aufbau und die Bereitstellung von Pipelines zu optimieren. Am Ende gehen sie mit sofort umsetzbaren Erkenntnissen nach Hause, die das Datenladen effizienter, smarter und sogar überraschend unterhaltsam machen.

Your SQL Database Is an MCP Server Now en

What if your SQL database could become an AI-ready platform without custom connectors, plugins, or risky text-to-SQL solutions? In this demo-driven session, you'll discover how Data API Builder and MCP turn SQL Server and Azure SQL into secure, governed endpoints for AI agents. See multiple AI clients connect to the same backend, enforce permissions, and deliver real business value through an open standard designed for the next generation of intelligent applications.

Copilot Is Not Copilot: What GitHub Copilot's Free Tier Actually Gets You en

"Copilot" isn't a product. It's a label Microsoft has stuck on more than eighty different things by now: a Copilot in Windows, one buried in Word, Excel, and Teams, a Security Copilot, a Copilot Studio for building your own agents, and somewhere in there, GitHub has its own Copilot too. These aren't variations on a theme. They're separate products built by separate teams for separate jobs, and most of the confusion inside IT departments right now comes from treating them as one thing.

We'll clear that up in the first few minutes: what each Copilot actually does, and who it's actually for. Then we drop the other seventy nine and spend the rest of the hour on the one that matters if you write code, GitHub Copilot, and specifically the tier most developers never look at twice: free.

Here's why that's worth your time. In June 2026, GitHub quietly rebuilt the whole pricing model around credits. Code completions and Next Edit suggestions are now unlimited, full stop, on every plan including free. Chat, agent mode, and premium models draw against a monthly credit allowance instead. That's not the free tier you tried in 2024 and shrugged off. It's close to a different product wearing the same name.

So we'll test it. No paid plan, no upsell, just VS Code and the free tier: chat for explaining and fixing real code, agent mode let loose on a small multi file task with nobody holding its hand. You'll walk out knowing exactly where free earns its keep, and exactly where it taps out.

Agent Showdown: GitHub Copilot vs Claude Code en

Same task, same starting query, same timer, two different AI coding agents. This session is a head to head, purely in T-SQL: GitHub Copilot and Claude Code each get an identical, bounded data question against a real small database, run one after the other, live, with the clock visible the whole time. Each round's timer is sized to its task, a short one for a simple write up, a longer one for a genuine diagnosis, but always identical between the two tools.

Each round is short and self contained on purpose, so a slow or wrong attempt is not a disaster, it is the result. Two rounds: write a new report query that genuinely spans several tables from a business question, and fix a query that quietly returns the wrong numbers because of a classic star schema join mistake. After each round we look at what actually got written or fixed, whether it finished inside the timer, and whether the result is something you would trust.

Both rounds run against a small, synthetic Oktoberfest sales database, tents, waiters, sales, food, so the queries are real and the questions being answered are at least entertaining. No cherry picked runs and no do overs, whatever each agent produces before its timer runs out is what the room sees and judges.

Why Choosing Between SQL Server and Fabric Is the Wrong Question de en

Many organizations approach modernization as a choice: stay on SQL Server or move to Microsoft Fabric. In reality, the most successful data platforms embrace both. This session explores practical architecture patterns that combine SQL Server, Azure SQL, and Fabric to deliver analytics, AI, and governance without forcing a full migration. Through real-world scenarios and reference architectures, you'll learn when each platform shines, how they work together, and how to build a future-ready data strategy.