How to Think Like the SQL Server Engine
Brent Ozar's free course that explains how SQL Server uses clustered and nonclustered indexes, why covering indexes help, and how to read the basics of an execution plan, using printed pages as a mental model. About two hours of video, with a written version too.
Course
Web Development
Brent Ozar Unlimited
About 2 hours
Intermediate
Free
What it is
How to Think Like the SQL Server Engine is a free course by Brent Ozar, a well-known SQL Server performance consultant and trainer. It explains, with a simple physical model of index pages, how SQL Server finds rows: the difference between clustered and nonclustered indexes, why and when a covering index helps, what key lookups cost, and the fundamentals of reading an execution plan. As of 2026-09 it is offered as a video version of about two hours and a written version as a series of articles, plus a follow-on "Big Data" edition for those who have finished the first (source: brentozar.com).
Is it worth your time?
Yes, if you use SQL Server and want to understand why some queries are slow. It is one of the clearest introductions to indexing available, and it builds the intuition that makes execution plans readable. It suits developers who can already write queries but have never designed an index.
Prerequisites
Comfortable writing SELECT queries with WHERE, joins and ORDER BY. No administration experience is needed.
Free vs paid
The course is free, but you register on Brent Ozar's training site to watch it. His other, longer courses are paid; you do not need them to benefit from this one.
How to use it well
- Follow along in your own SQL Server with the demo database the course uses, and run each query with the actual execution plan and STATISTICS IO on.
- After each section, predict the plan and the number of logical reads for a query before you run it.
- Afterwards, apply the ideas to one slow query in your own project and write down the before and after numbers.
What you can do after it
Explain clustered versus nonclustered indexes, decide when a covering index is worth it, spot key lookups in a plan and reason about why SQL Server chose a scan or a seek.
Common questions
Does it apply to other databases? The ideas of B-tree indexes and covering indexes carry over, but the details, such as clustered tables, are specific to SQL Server. For PostgreSQL and MySQL, Use The Index, Luke covers similar ground.
Is it current? Index fundamentals change slowly; the core lessons apply to current SQL Server versions.
Ready to start?
Opens on Brent Ozar Unlimited in a new tab.
Stuck on something specific?
Writing only gets you so far. If you want an answer to your situation rather than the general case, book a session and we will work through it together. Sessions are free for approved Sefism members, and a few slots open each week.
Follow along
New writing, resources and project ideas land here first.