Solution Architects and DBAs, responsible for the implemented architecture.
The incredible Columnstore Indexes can increase your analytical query processing speed multiple times, they are updatable (Clustered from SQL Server 2014 and Nonclustered from SQL Server 2016 respectively), but they keep on supporting different sets of the functionalities – such as Change Data Capture (Nonclustered Columnstore) or LOBs (Clustered Columnstore), and this brings a great confusion onto the table.
This session will light up your path on when to use what functionality to use and when, even though sometimes one of the type of the Columnstore Indexes does not seems to appear as a default choice for your scenario.
Why I Want to Present This Session:
There is a great confusion on when to choose which Columnstore Index, and having faced real-world implementations with the “wrong” Columnstore Indexes, I wanted to help everyone to discover the right solution.
Also, given the limitations and supported features, I see an incredible need to explain which features can be used in which architectures (think HA/DR scenarios, HTAP reporting scenarios, etc)
Latest posts by Niko Neugebauer (see all)
- Columnstore Clustered vs Nonclustered, when to choose what - December 27, 2017
- Worst Practices & Less Known Limitations for Columnstore Indexes - April 21, 2017