Margarita Naumova
CEO, Data Platform Architect at Inspirit
Oslo, Norway
Actions
Magi Naumova is an SQL Server Architect and Consultant, speaker and trainer, Microsoft Certified Master, MVP, MCT, Founder of SQL Master Academy, and founder and CEO Of Inspirit Data Platform Consulting Company in Bulgaria and Norway, founder, and the leader of Bulgarian SQL & BI User group and Azure Analytics User Group BG. She has more than 25 years SQL Server training and consulting experience. Magi is a former member of Microsoft Services Worldwide Technical Leadership Team. Currently she speaks, writes, trains and consults people on SQL Server and MS Fabric Data Platform. She is MVP for 15 years in a row. Her unique SQL Master Academy training program has helped hundreds of specialists to feel knowledgeable in their daily work or to find an inspirational career path in the world of SQL Server.
Badges
Area of Expertise
Topics
A Deep Dive into Table Partitioning – Part 2: Design, Maintenance, and Troubleshooting
Table partitioning isn’t a silver bullet for large tables. Used well, it improves data management and maintenance; used poorly, it leads to unpredictable performance and difficult troubleshooting.
In this two-part, demo-driven deep dive, we build a clear understanding of how table partitioning works in SQL Server and how design and operational choices affect performance. The sessions move from partitioning mechanics and elimination to real-world design, indexing, and troubleshooting, giving attendees a practical mental model they can apply confidently in production.
Part 2 builds on the mechanics covered in Part 1 and focuses on real-world design and operational challenges. We examine partition key selection, aligned and non-aligned indexes, and how these choices affect both query performance and maintenance.
Using practical scenarios, we explore common performance problems in partitioned tables, why “partitioning made it slower,” and how to troubleshoot these issues effectively. The session closes with a practical checklist for diagnosing and fixing partitioned-table performance problems in production systems.
A Deep Dive into Table Partitioning – Part 1: Mechanics and Performance
Table partitioning isn’t a silver bullet for large tables. Used well, it improves data management and maintenance; used poorly, it leads to unpredictable performance and difficult troubleshooting.
In this two-part, demo-driven deep dive, we build a clear understanding of how table partitioning works in SQL Server and how design and operational choices affect performance. The sessions move from partitioning mechanics and elimination to real-world design, indexing, and troubleshooting, giving attendees a practical mental model they can apply confidently in production.
Part 1 focuses on the mechanics of table partitioning and how SQL Server actually uses partitions. We start with partition functions and schemes, how boundary values are interpreted, and how rows are positioned across partitions—foundations that are critical for designing partitioning correctly.
Through practical demos, we explore core partition operations such as switching data in and out, splitting and merging partitions, and the mechanical constraints these operations introduce. With this foundation in place, we move into partition elimination: what enables it, what breaks it, and how to recognize effective pruning by reading execution plans.
SQL Database in Microsoft Fabric: What It Is, How It Works, and Why It Matters
SQL Database in Microsoft Fabric is not “just another database”—it’s a reimagined, cloud-native relational engine built directly into the Fabric ecosystem. But what does that mean in practice? In this session, we’ll demystify SQL Database in Fabric by covering:
• What it is: How SQL Database in Fabric differs from traditional Azure SQL or SQL Server
• How it works: Behind-the-scenes architecture, scalability model, and integration with other Fabric components (e.g., Lakehouse, Pipelines, Real-Time Hub)
• Why it matters: Use cases that benefit from a native SQL layer inside your analytics platform—like real-time ingestion, metadata consistency, and low-latency querying
We’ll also discuss limitations, patterns, and coexistence with traditional databases. Expect practical demos, architectural guidance, and honest discussion of where SQL Database in Fabric fits in a real-world data platform strategy.
SQL Server 2025 Meets Microsoft Fabric: Seamless Integration for Modern Data Solutions
Explore the powerful integration between SQL Server 2025 and Microsoft Fabric, designed to revolutionize modern data management and analytics. This session will highlight how SQL Server seamlessly connects with Fabric’s unified data platform, enabling advanced analytics, data engineering, and real-time insights. Learn how to leverage this integration for streamlined data workflows, enhanced scalability, and simplified governance. Whether you're building data pipelines, optimizing queries, or enabling cross-platform analytics, this session will provide practical strategies and demos to help you unlock the full potential of SQL Server 2025 and Microsoft Fabric.
SQL Server 2025: Inside the Database Engine Performance Improvements
SQL Server 2025 introduces a set of focused performance enhancements aimed at improving execution speed, concurrency, and resource efficiency. In this session, we’ll dive into the latest capabilities in the Database Engine, including Intelligent Query Processing improvements (adaptive joins, memory grant feedback), better TempDB scalability, and optimizations for columnstore indexes and Batch Mode on Rowstore. We’ll also look at support for persistent memory and how these changes impact both OLTP and analytical workloads. This session is designed for DBAs and developers who want to understand not just what’s new, but how to apply these features effectively in production scenarios.
SQL Server 2025: The One-Day Masterclass
SQL Server 2025 introduces some of the most exciting changes in the history of the platform — from AI integration and vector search to enhanced performance, streaming, and modern T-SQL. This full-day masterclass is designed to give developers, DBAs, and data engineers everything they need to hit the ground running with the latest release.
Across one packed day, we’ll explore:
Engine & Performance Enhancements for DBAs
Learn about query processing improvements, advanced Query Store capabilities, optimized locking, memory grants, and new availability options that make SQL Server 2025 faster and more reliable than ever.
T-SQL & Developer Innovations
Discover modern T-SQL features, including JSON and regex support, improved error handling, REST/HTTP integration, and new capabilities that make coding against SQL Server more productive and powerful.
AI, Vector Search & Semantic Capabilities
See how SQL Server 2025 brings intelligence inside the database with built-in vector search, embeddings, and semantic search, enabling AI-driven applications without leaving your data platform.
Data Engineering & Real-Time Streaming
Explore change event streaming, real-time analytics integration with Microsoft Fabric, and how SQL Server 2025 fits into modern hybrid and cloud-first data architectures.
Migration, Best Practices & Roadmap
Wrap up with guidance on upgrading, modernizing, and aligning your SQL Server strategy to take full advantage of everything 2025 has to offer.
Whether you’re a developer looking to build AI-powered apps, a DBA focused on performance and availability, or a data engineer driving real-time insights, this pre-con will equip you with the knowledge and practical skills to make SQL Server 2025 work for you.
SQL Server 2025: Unleashing Next-Level Database Performance
Discover how SQL Server 2025 takes database performance to new heights with cutting-edge enhancements to the Database Engine. This session dives into the latest innovations, including improved query processing, enhanced scalability, and optimized storage capabilities. Gain insights into real-world use cases, performance benchmarks, and practical tips to maximize the power of SQL Server 2025 for your organization. Whether you're a DBA, developer, or IT professional, this session will equip you with the knowledge to supercharge your database environment.
Table Partitioning DOs and DONTs
Dealing with billion rows tables in SQL Server is challenging. Resolving performance issues when working with them is not easy either. Sometimes we consider implementing table partitioning to try to deal with large tables issues. But table partitioning, just the same as indexing, is not a silver bullet. There are few important things we need to consider before implementing partitioning. In this demo-rich session we will cover these key points. We will understand how the partitioning works and how it helps for better data management and maintenance. When do we get improvement in the performance, and when should we better not choose partitioning but some other optimization techniques instead. Shall we divide our large tables in partitions, or we could just create yet another index to improve query response time? These are just some of the questions you will find answers in this session.
Warehouse loading – tips and tricks for better performance
Microsoft Fabric offers built-in data ingestion tools that allow users to ingest data into warehouses at scale using code-free or code-rich experiences. In this session we will discover all the possible options - T-SQL features, and code free tools, and even more that some source types provide.
We will compare them according to different criteria like performance and ease of use, source volume and types, target schemas, single or cross workspaces usage. You will see them in action and understand how they work. You will get some tips on their best usage and their best combinations. But more importantly you will understand the main and the most critical factors for the design of the loading process so that you can make it fast and smooth. We will talk about how to combine the toolset we have for building elegant data loading solutions in the warehouse for the best performance. Tips and tricks on the way will help you learn, design and be prepared to start a process of loading data in your Fabric warehouse environment.
When Indexes Can’t Save You: Query Patterns That Need Refactoring
Not every performance problem can be fixed with an index. In this session, we’ll look at common query and schema patterns where indexing falls short—and what to do instead. You’ll learn how to recognize anti-patterns, refactor for better performance, and design queries and tables that work with the optimizer, not against it.
Warehouse performance tuning from loading to querying
Monitor, analyze and optimize your warehouse workload in Fabric!
Understand the warehouse query processing and query optimization and get the best practices for working with the warehouse from loading to querying! During this session we will go through a sample demo solution all the way from data source to querying and usage of a warehouse.
You will see how to monitor, baseline and optimize the workload at every stage on the way so that it runs efficiently and uses the least possible compute resources. We will discover best practices on how to organize your loading process, how to baseline and optimize it. You will understand query processing and query optimization of the warehouse engine, the delta store and the statistics. You will see the toolset we can use for baselining and monitoring out-of-the box. This will cover the Fabric Monitor and Fabric Capacity Metrics, query insights, and common DMVs and you will learn how to read the output and use them in different cases. We will make some comparisons with Lakehouses in Fabric and with other Data Stores outside of Fabric as well.
Whether you use warehouse for gold layer or as a main data store in Fabric, it needs a deeper dive and discovery. This session will give that to you in the best possible way – by example.
Performance tuning with Copilot. Are we there yet?
Performance tuning in Azure is critical and at the same time not an easy task. It is important we put effort and time in it as proper optimization will reduce the resource consumption and cost. With the release of Copilot for Azure SQL, we got a new and promising way to make this effort easier and to get the answers to our questions fast. It sounds tempting to be able to get some copilot-ed help for questions like where the major cost spending is, why and how to reduce it, what are the active connections running right now or what are the top high CPU queries run in the last week? Even if it’s tempting, we still have some doubts using it, is it useful or it is just a funny toy to play with? Shall we trust it or its just yet another fancy tool out there? Come to this session and let's experience together performance tuning with Copilot and let's see what we can rely on and what to expect.
Next-Level Data: Exploring SQL Server 2025 Innovations
SQL Server 2025 is more than just an update — it's a leap forward in how we store, process, and extract value from data. In this session, we’ll dive into the most impactful new features and enhancements that make this release a game changer for data professionals. From native vector data types and AI-powered search to performance optimizations, enhanced security, and cloud-connected capabilities, you'll get a guided tour of what's new, why it matters, and how to take full advantage of it in your environment. Whether you're a DBA, developer, or architect, this session will equip you with practical insights and real-world examples to help you level up your SQL Server skills for 2025 and beyond.
Next generation Data Warehousing with Microsoft Fabric
Microsoft Fabric provides customers with a unified product that addresses every aspect of their data estate by offering a complete, SaaS-ified Data, Analytics and AI platform, which is lake centric and open. What does it mean for Data warehousing and how it changes the game in the BI and Data Engineering space? Come to this session to share the exciting new feature of Cloud Analytics with me. I will talk about Lakehouse, Datawarehouse, Data Mesh and much more on what I think will be part of our job next few years.
Next generation Data Warehousing with Microsoft Fabric
MS Fabric has opened a whole new world of data warehousing! The change is not just about the technology; it’s a paradigm shift in the data warehouse field. Come to this session to share the exciting new feature of Cloud Data warehousing. Let’s see how the next generation data warehouse looks like and what will change with the release of MS Fabric. We will discover the Fabric Datawarehouse, Lakehouse, and how to cross them with shortcuts and queries. You will see how democratizing and SaaS-ifying the Datawarehouse opens a whole new set of scenarios, and how to leverage a Lakehouse together with a Warehouse for a best of breed analytics strategy!
MS Fabric & Databricks - Happily ever after or Clash of the Titans
What is happening in the field of Data Engineering and what should we expect in the future? Which platform for data engineering to focus on if you are Data Engineer?
Databricks is a cloud-based data processing platform that provides a collaborative environment for data scientists, engineers, and analysts. On the other hand, Fabric is a unified analytics platform that brings together all the data and analytics tools that organizations need. Will this be a battle or a good working union? Let’s explore the key differences between Microsoft Fabric and Databricks from many angles including technical specifics and capabilities, experience from projects so far and even pricing. Let’s also talk about use cases of both platforms.
Lakehouse vs Data Warehouse vs KQL Database: Use Cases and Architecture Designs
What architecture designs and solutions are best for analytics in Fabric, what to choose for building your solution – Lakehouse, Data Warehouse or KQL Database and where to start? You will find all the answers during this training day! We will dive into all 3 items - Lakehouse, Data Warehouse and KQL Database to learn their specifics and their best usage. We will emphasize the main differences and talk about typical uses cases. Then we will pay a special attention to common Architecture Designs for using Data Warehouse, Data Lakehouse, and Real-Time Analytics/KQL Database. At the end you will discover how your data estate for analytics data warehousing/reporting will change or differ from existing designs and how to choose the right path moving forward. This training day will give you a very good understanding of the differences between the Data Warehouse, Data Lakehouse, and KQL Database, and most important it will explore for you possible Fabric solution designs and use cases to get the best of the Data Warehouse, Data Lakehouse, and Real-Time Analytics/KQL Database. Isn’t it that you need to know to start or continue your journey to a Modern Analytics Platform in Fabric as data engineer or data architect?
Lakehouse vs Data Warehouse vs KQL Database: Use Cases and Architecture Designs
What architecture designs and solutions are best for analytics in Fabric, what to choose for building your solution – Lakehouse, Data Warehouse or KQL Database or a combination of them? How and where to start? This session will give you the answers and even more. We will dive into all 3 experiences - Lakehouse, Data Warehouse and KQL Database and discover their best usage. We will emphasize the main differences and talk about typical uses cases. Then we will pay a special attention to common Architecture Designs for using Data Warehouse, Data Lakehouse, and Real-Time Analytics/KQL Database. At the end you will discover how to choose the right path moving forward. Isn’t it that you need to know to start or continue your journey to a Modern Analytics Platform in Fabric as data engineer or data architect?
Keeping historical data in tables forever – mission (im)possible!
Data growth can easily become a problem soon after deploying a database in production, especially if data layout and data management was not planned. Let’s discover some solutions for keeping historical data in the database when you receive near to impossible requirements like storing data in same tables forever and being able to edit and query them at the same time, of course keeping the response at its best. Will combining partitioning, temporals, CSI, and other indexing with different loading/unloading/deletion make it possible? The session starts by defining the frame requirements coming from real projects and goes through different alternatives and possible solutions, their benefits, and drawbacks. Based on a real project case the session walks you through the design process from the start to the reaching of the final solution and making the client (and developers) happy.
GenAI in SQL Server 2025: What Works, What Doesn’t, and Why It Matters
SQL Server 2025 introduces native support for vectors, AI model integration, and RAG workloads—but what does this mean for system design, performance, and scalability? This session takes a critical look at the architectural implications of using SQL Server for GenAI applications. We’ll explore how vector indexes work, how embedding data affects performance, and when SQL Server is (or isn’t) the right tool for the job. Ideal for architects and DBAs looking to make informed decisions in AI-powered data platforms.
This session is for practitioners who want to go beyond the buzzwords and understand what it takes to use SQL Server effectively in AI-powered applications—what works, what’s still experimental, and what to watch out for.
From Load to Logic: Designing for Performance in Microsoft Fabric Warehouses
Microsoft Fabric Data Warehouse is not just another cloud database — it’s the T-SQL analytics backbone in Azure and the strategic destination for modernizing on-premises SQL-based data warehouses. This session will dive deep into designing for performance and scalability in Fabric, equipping you with the patterns, techniques, and real-world practices to get the most out of this SaaS-based solution. This session is your guide to mastering the SQL path in Fabric: where familiar skills meet enhanced cloud-native performance.
Whether you're migrating from an on-premises SQL Server or Synapse Dedicated SQL Pools, understanding how to organize your data loads, structure your models, and tune your workloads is critical. We'll explore enhanced query performance features such as result set reuse, automatic statistics, query plan caching, and intelligent workload management — all designed to give you cloud-grade agility without giving up the control you expect from traditional systems.
Learn how to navigate limitations, optimize for throughput and concurrency, and make the most of Fabric's underlying resource management. This session is your guide to aligning enterprise-grade SQL data warehousing with the simplicity and scale of the cloud.
Designed for professionals who live and breathe SQL, this session shows how to bring enterprise-grade data warehousing into the Fabric era — without sacrificing control, performance, or architectural clarity.
A Table Partitioning Deep Dive - Part 2
From zero to hero, understand table partitioning now and forever! Partitioning is not a silver bullet of our biggest tables! Many thinks can go wrong. Before we consider it, we need to understand how it works. In this demo-rich session we will cover very much of what do you need to know about it. In the Part 1 of this 100 min session we will start from table partitioning mechanics and will dive very fast into mastering it. We will cover partition elimination, aligned and non-aligned indexes. In the Part 2 we will pay special attention on choosing the partition key, will touch a bit reading query plans of partitioned tables, and troubleshooting performance problems. At the end of this deep dive session, you will get a very good understanding of how the partitioning works and how it helps for better data management and maintenance. On top of technical demos, I will give you examples from many real cases I’ve gone through. You will get ideas on how to approach more complex cases of historical data maintenance, by combining partitioning, indexing, and records deletion. You will know how to achieve performance improvement, and when you should not choose partitioning but some other optimization techniques instead.
Ask the Expert - Group A (Wed 11:10 to 13:00)
'Got a problem? Don't know who to call? Bring it to SQLBits and get it solved by our panel of experts!
More information can be found here:
https://sqlbits.com/news/ask-the-experts-at-the-experts-lounge/
1. SQL and Databases:
2. Microsoft Fabric and Related Technologies:
3. Power BI and Data Visualization:
4. Azure and Cloud Technologies:
11:10 to 12:00
1. Magi Naumova
2. Paul Andrew, Brynn Borton
3. James Dales, Ana-Maria Bisbe York
4.
12:10 to 13:00
1. Martin Catherall & Heid Hasting
2. Freddie Santos, Nagaraj Sengodan
3. Kristoffer West, Mark Hayes
4.
Approaching the biggest tables in your database – strategies and best practices
Ever-growing multimillion rows tables are challenging to work with. It’s not about the storage, it’s about keeping the performance of your workload even when data grows forever. What options do we have? Should we use partitioning, deletion and archiving, or some indexing and temporals? When to use what and how? Let’s discover some strategies for data maintenance before it becomes even more difficult to implement them.
A Data Lakehouse walkthrough with Synapse Analytics
Discover the Lakehouse concept as a potential roadmap of your Modern DW in the cloud.
During this session we will go through creating different types of objects in Synapse Serverless, combining them in a database-like solution, creating delta format in Spark and using it in a Lake database in Serverless. If you speak T-SQL and understand ADL, then this session will help you to level up fast in the data engineering track by understanding Synapse Serverless and the concept of Data Lakehouse in practical examples. You will learn some tips and trick on the way, and I promise I will not speak Phyton (too) much!
Data Lakehouse with Azure Synapse Analytics
Synapse Analytics combines the use of relational datawarehouse and a big data ADL in one single workspace. Data lakehouse on the other hand is the ability to query data directly and in the same way no matter if the data are in data lake or relational datawarehouse. Is it just the workspace that makes the concept of Data lakehouse real in Synapse Analytics? Let discover what we can do with Synapse serverless, how to architect data analytics in the future and do these new trends come to say that the traditional DW is dead?
Azure SQL Managed Instance - what, how, when (and how much)
Do you want to learn more about SQL MI? Then this session is a good start. I will show you how it looks like, how to migrate your databases to MI and of course I will talk about the price. I believe some common use case scenarios will pop up and you will be able to plan your path to SQL MI.
Modern database design (anti)patterns
The database tier is quite often neglected in the application design process, which results in performance issues, lack of scalability and even disruption of service. Design choices that seem great at the beginning fail totally on scale and performance when the application goes in production. We must realize that patterns which were valid 10 years ago are less likely to work now, like cursor logic, xml usage, or storing all in db v/s using NoSQL. From global industry trends to specific database patterns, this session is a combination of best practices, good and bad patterns, tips, and tricks which I give to customers in my work as a consultant. Good design choices from the beginning help avoiding complex and expensive redesign in the future!
feedback link: https://sqlb.it/?7315
SQLBits 2026 Sessionize Event
SQL Konferenz 2026 Sessionize Event
Fabric February 2026 Sessionize Event
MVP-Dagen 2025 Sessionize Event
Data Saturdays Sofia 2025 Sessionize Event
European Microsoft Fabric Community Conference 2025 Sessionize Event
Data Saturday Gothenburg 2025 Sessionize Event
Data Saturday Oslo 2025 Sessionize Event
SQLBits 2025 - General Sessions Sessionize Event
Discovery Day 2025 Sofia Sessionize Event
Global Azure Bulgaria 2025 Sessionize Event
Data Saturdays Sofia 2024 Sessionize Event
Data Saturday Oslo 2024 Sessionize Event
Data Saturday Stockholm 2024 Sessionize Event
Global Azure Bulgaria 2024 Sessionize Event
SQLBits 2024 - General Sessions Sessionize Event
Fabric February 2024 Sessionize Event
Data Saturday Sofia 2023 Sessionize Event
Data Saturday Oslo 2023 Sessionize Event
Global Azure Bulgaria 2023 Sessionize Event
Data Saturday Stockholm 2023 Sessionize Event
SQLBits 2023 - General Sessions Sessionize Event
Data Saturday Sofia 2022 Sessionize Event
Data Saturday Oslo 2022 Sessionize Event
Data Saturday Plovdiv 2022 Sessionize Event
Global Azure Bulgaria 2022 Sessionize Event
SQLBits 2022 Sessionize Event
Data Saturday Sofia Sessionize Event
Data Saturday Plovdiv Sessionize Event
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top