Kevin Feasel

SQL Server 2025 reshapes the edition landscape. Standard Edition gains features that previously required Enterprise. Developer Standard Edition is a new option for dev/test environments. Web Edition is gone entirely. These changes affect licensing costs, architecture decisions, and migration...

AI is being pulled into the data layer faster than most organizations can articulate why. The pressure is applied into every roadmap, every vendor conversation, and every leadership meeting now includes some version of ""what's our AI story?"" The proximity to data is not the same as...

DuckDB is a lightweight, open-source database that has become a go-to solution for running analytical queries on local machines. In this talk, we will explore why that is, showing off some of its simplicity, portability, and extensibility.

The Fabric Eventhouse is a powerful data store that classical T-SQL developers may miss out on. Using the Kusto Query Language (KQL), we can quickly and efficiently parse and slice data, and operate over time series.

Data assets at organizations tend to stick around for a while--many of the databases we work with are old enough to vote. These databases tend to be full of code and patterns which made sense in SQL Server 2005 but are not the best approach today.

DuckDB is a lightweight, open-source database that has become a go-to solution for running analytical queries on local machines. In this talk, we will explore why that is, showing off some of its simplicity, portability, and extensibility.

DuckDB is a lightweight, open-source database that has become a go-to solution for running analytical queries on local machines. In this talk, we will explore why that is, showing off some of its simplicity, portability, and extensibility.

Transact-SQL (T-SQL) is a powerful language for querying data but it can be a challenge for application developers and business analysts who are not already deeply familiar with the language to solve business problems.

Regular expressions, a powerful technique for matching arbitrary strings, are now available in SQL Server 2025. In this session, we will review how regular expressions work in SQL Server.

DuckDB is a lightweight, open-source database that has become a go-to solution for running analytical queries on local machines. In this talk, we will explore why that is, showing off some of its simplicity, portability, and extensibility.

Natural language processing is a vital field within data science and machine learning that combines computer science and linguistics to analyze and generate text in the languages we speak.

Have you ever worked on a project where you wanted a quick and easy way to interact with and visualize data? This is exactly the use case for Streamlit, a Python library intended to build data applications with ease.

Natural language processing is a vital field within data science and machine learning that combines computer science and linguistics to analyze and generate text in the languages we speak.

DuckDB is a lightweight, open-source database that has become a go-to solution for running analytical queries on local machines. In this talk, we will explore why that is, showing off some of its simplicity, portability, and extensibility.

Forensic accountants and fraud examiners use a range of techniques to uncover fraudulent journal entries and illegal activities. As data professionals, most of us will never unravel a Bernie Madoff scheme, but we can apply these same techniques in our own environments to uncover dirty data.

PolyBase is Microsoft's primary data virtualization technology, allowing you to integrate SQL Server with a variety of external systems such as Oracle, MongoDB, Cosmos DB, Teradata, Apache Spark, and even other SQL Server instances. In this talk, we will walk through the mechanics of PolyBase, showing how we can write…

Your organization's data will always be a primary target for adversaries. If you are using SQL Server to store this data, there are a wealth of settings and tools available to improve your security posture.

Transact-SQL (T-SQL) is a powerful language for querying data but it can be a challenge for application developers and business analysts who are not already deeply familiar with the language to solve business problems.

Natural language processing is a vital field within data science and machine learning that combines computer science and linguistics to analyze and generate text in the languages we speak.

You've tried Azure OpenAI and played around with its Large Language Models and image generation with Dall-E. Now your product team wants you to incorporate generative AI into your existing .NET applications.

PolyBase is Microsoft's primary data virtualization technology, allowing you to integrate SQL Server with a variety of external systems such as Oracle, MongoDB, Cosmos DB, Teradata, Apache Spark, and even other SQL Server instances.

Discovering outliers is a critical part of systems monitoring, financial auditing, and more. In this full-day workshop, we will gain an understanding of what constitutes an anomaly or an outlier, starting with a general concept of the terms and then moving into technical definitions.

The major cloud providers--such as Amazon's AWS, Microsoft's Azure, and Google's GCP--each offer dozens of separate services. These services can be quite useful, but sometimes understanding just how much they cost can be difficult.

Have you ever worked on a project where you wanted a quick and easy way to interact with and visualize data? This is exactly the use case for Streamlit, a Python library intended to build data applications with ease.

Product demos can make or break a sale, yet technical sales specialists often accidentally miss the mark by introducing unnecessary complexity into explanations, turning what should be an easy explanation into an overwhelming slog. This talk covers Peter Cohan's Great Demo!

Azure Machine Learning provides extensive capabilities for low-code and full-code machine learning tasks. Once you have a full-code solution in place, the next step is to take advantage of Machine Learning Operations (MLOps).

Data assets at organizations tend to stick around for a while--many of the databases we work with are old enough to vote. These databases tend to be full of code and patterns which made sense in SQL Server 2005 but are not the best approach today.

PolyBase is Microsoft's primary data virtualization technology, allowing you to integrate SQL Server with a variety of external systems such as Oracle, MongoDB, Cosmos DB, Teradata, Apache Spark, and even other SQL Server instances.

Azure Machine Learning offers a low-code approach to performing machine learning tasks, but what happens when you need to do something a little more complex? In this talk, we will go beyond the Azure Machine Learning designer and show how to use Azure Machine Learning in a code-first environment.

Window functions are a powerful technique for analyzing data in SQL Server. In this session, we will understand when to use a window function like ROW_NUMBER() or LAG(), including specific use cases.

Anomaly detection helps us find data points which are inconsistent with the rest of a dataset. But what does this mean when our data points have many individual features?

Negotiating a downtime for a release can be difficult in this modern, 24x7 world. When downtime means lost revenue, the price of taking a downtime to deploy code can run up quickly.

One of the most important tasks for a database administrator is taking (and testing!) backups. As databases get larger and larger, the amount of time it takes to perform a backup can grow as well, to the point where your backups take longer than your available backup window.

Negotiating a downtime for a release can be difficult in this modern, 24x7 world. When downtime means lost revenue, the price of taking a downtime to deploy code can run up quickly.

Negotiating a downtime for a release can be difficult in this modern, 24x7 world. When downtime means lost revenue, the price of taking a downtime to deploy code can run up quickly.

Over the past several years, we have seen a resurgence of interest in functional programming languages, particularly with the success of Scala and R in the marketplace.

Negotiating a downtime for a release can be difficult in this modern, 24x7 world. When downtime means lost revenue, the price of taking a downtime to deploy code can run up quickly.

Negotiating a downtime for a release can be difficult in this modern, 24x7 world. When downtime means lost revenue, the price of taking a downtime to deploy code can run up quickly. This talk will show you a way to reduce the cost of deployment by reducing or even eliminating required downtime for database code deployment. We will show how to release new database code, update existing stored procedures, refactor and restructure tables, and even perform systemic re-architecture with no downtime and minimal blocking. We will use as a case study one release in which we modified the primary key column on 180 tables and updated over 550 stored procedures without end users noticing a thing. If you want to hit that three nines (or better) service-level agreement, this talk will help get you on the road to reduced downtime.

Forensic accountants and fraud examiners use a range of techniques to uncover fraudulent journal entries and illegal activities. As data professionals, most of us will never unravel a Bernie Madoff scheme, but we can apply these same techniques in our own environments to uncover dirty data.

Writing code which uses SQL Server Machine Learning Services is easy enough, but that is just the beginning of the story. In this talk, we will cover the rest of the R development story for database administrators.

Over the past several years, we have seen a resurgence of interest in functional programming languages, particularly with the success of Scala and R in the marketplace.

The Naïve Bayes series of algorithms are some of the simplest classification algorithms, but they tend to offer reasonably good results very quickly for a number of problems, including Natural Language Processing problems such as spam classification, as well as more classical feature-driven classification.

Microsoft introduced the APPLY operator in SQL Server 2005. This under-used, non-standard operator can help us solve a number of common (and not-so-common) problems with elegant code which performs well.

As companies work to gain insight from ever-increasing amounts of data, data platform practitioners need tools which can scale along with the data. Early big data solutions in the Hadoop ecosystem assumed that data sizes overwhelmed available memory, emphasizing heavy disk usage to coordinate work between nodes.

There's an old adage in software development: Garbage In, Garbage Out. This adage certainly applies to data science projects: if you simply throw raw data at models, you will end up with garbage results.