SQL Server statistics

Question

I have run the Database Tuning Advisor and it has suggested that some statistics are created. Whilst this is great and I understand the purpose of statistics, when I have reviewed them it has created 5 or more with very similar columns included in them. If I went ahead and created these statistics, if a column is replicated in more than one of these statistics and the first statistic is updated, how does this update affect the other statistics?

Thanks,

Answer 1

Including multiple columns in statistics adds a new dimension that cannot be estimated otherwise: correlation.

• Query Predicate Contains Multiple Correlated Columns

  When a query predicate contains multiple columns that have cross-column relationships and dependencies, statistics on the multiple columns might improve the query plan. Statistics on multiple columns contain cross-column correlation statistics, called densities, that are not available in single-column statistics. Densities can improve cardinality estimates when query results depend on data relationships among multiple columns.

• Optimizing Queries That Access Correlated datetime Columns

• Exploring Column Correlation and Cardinality Estimates

, if a column is replicated in more than one of these statistics and the first statistic is updated, how does this update affect the other statistics

All statistic need to be updated to be accurate.

Whether all stats recommended by the DTA are necessary, that is a different topic.