Prerequisites Before the Class
Hi, Brent here! I’m really excited to welcome you to Fundamentals of Columnstore Indexes.
Watch How to Think Like the Engine and Fundamentals of Index Tuning.
Before we get started, I need to make sure that we’ve covered the basics:
- How to Think Like the Engine – free 90-minute class that covers the foundations of clustered and nonclustered indexes, statistics, memory grants, and sargability.
- Fundamentals of Index Tuning – one-day class that covers how to index for WHERE, ORDER BY, and JOINs, and covers how to pick the field order for indexes.
- If you’re wondering why we only focus on SQL Server 2016 & newer in this class, and your production servers are on 2014 or older, read the documentation on what’s new in columnstore indexes, by version.
Armed with the material in those two classes, you’ll be well-equipped to keep up in this class.
If you're attending a live class, join Slack.
To ask questions and talk to the other attendees, go to https://www.brentozar.com/slack/ to get an instant invite, then join the #BrentOzarUnlimited channel. This way attendees can help each other with their questions during the live labs.
Optional: set up your lab VM.
You don't have to do this: you can watch the entire class without having your own VM. This class doesn't have any homework exercises for you. This is only optional if you want to follow along and do the same scripts I'm doing.
In the rest of my Fundamentals classes, I use a small Stack Overflow database, but that doesn’t cut it for Fundamentals of Columnstore. Columnstore indexes are all about big data, so for this class, I use the same setup as my Mastering classes.
Read these instructions on how to set up a Mastering class VM. It'll take several hours. (The great news is that after setting your VM up for this class, you're good to go for the rest of my Mastering classes.)
After that, in one of the final modules of this class, we build a clustered columnstore index on the Votes table. It takes about 5-10 minutes. If you want to get that out of the way ahead of class, or on the morning of class, you can run the below demo script: