Statistics in SQL Server - Microsoft SQL Server 2012 Performance Tuning - page 294

3. Here is the screenshot of the statistics window generated by the previous script:

4. After observing the All density column of the density vector result set, let us drop the

index we created previously and recreate a non-clustered index with a filter:
DROP INDEX idx_WorkOrder_ScrapReasonID on [Production].[WorkOrder]
GO

CREATE INDEX idx_WorkOrder_ScrapReasonID on [Production].
[WorkOrder](ScrapReasonID)
WHERE ScrapReasonID IS NOT NULL
GO

5. Now, let us look at the All density column of the density vector result set, with the

following script:
DBCC SHOW_STATISTICS ("[Production].[WorkOrder]",idx_WorkOrder_
ScrapReasonID)
GO

DROP INDEX idx_WorkOrder_ScrapReasonID on [Production].[WorkOrder]
GO