Tag : cardinality - SentryOne Team Blog

Tag: cardinality

Paul White : Cardinality Estimation : Combining Density Statistics

Paul White : Cardinality Estimation : Combining Density Statistics

Paul White (@SQL_Kiwi) whips out the debugger while showing how SQL Server derives estimates when combining multiple, single-column statistics.

Paul White : Cardinality Estimation for a Predicate on a COUNT Expression

Paul White : Cardinality Estimation for a Predicate on a COUNT Expression

Paul White (@SQL_Kiwi) takes you on an optimizer journey, exploring how SQL Server comes up with cardinality estimates for COUNT queries.

Erin Stellato : Trace Flag 2389 and the new Cardinality Estimator

Erin Stellato shows the combined effect of different versions of the cardinality estimator and trace flag 2389 on an ascending key model.

Cool Features of SQL Server 2014: Improved Cardinality Estimates

COOL FEATURES OF SQL SERVER 2014: IMPROVED CARDINALITY ESTIMATES With all the cheerleading and the steady drumbeat of new features being released to Azure, it's easy to lose track of the many cool and valuable new features released in the on-premises version of SQL Server. One of the crown jewels of SQL Server, the cardinality

Statistics – What, why and how (Part 3)

Lori Edwards (@loriedwards) continues her series on SQL Server statistics, this time showing how even up-to-date statistics can yield inaccurate estimates.

Statistics – What, why and how (Part 2)

Lori Edwards (@loriedwards) continues her series on SQL Server statistics, explaining how the optimizer uses cardinality estimates.

Paul White : A Subquery Cardinality Estimation Bug

Paul White (@SQL_Kiwi) uses several examples to describe a bug in cardinality estimation that can be quite difficult to spot if you don't know where to look.

Aaron Bertrand : New Trace Flag to Fix Table Variable Performance

Following the recent release of SQL Server 2012 Service Pack 2, SQL Sentry's Aaron Bertrand (@AaronBertrand) takes a first look at a new performance enhancement and trace flag aimed at improving cardinality estimates for table variables and table-valued parameters.