Just a caution...

The WEEK (ww) datepart is ALWAYS based on Sunday for DATEDIFF.  MS did that to ensure that the DATEDIFF(wk) functionality will be "Deterministic" so that it can be...

SET DATEFIRST 6 ; -- Set first Day of Week to Saturday SELECT [CreateDateTime],DATEPART(WEEK,[CreateDateTime]) - DATEPART(WEEK,CAST(CAST(YEAR([CreateDateTime]) AS VARCHAR(4))+'-' + CAST(MONTH([CreateDateTime]) AS VARCHAR(2)) + '-01' AS DATETIME))+1 AS WeekNo FROM...

Hi

We have an stored procedure which runs frequently and  it contain logic in which data are first dumped in temproray table and then moved to physical table .due to...

It's real simple... have them prove that the performance is better with repeatable, demonstrative code.  One good test is worth a thousand expert opinions and until they can prove it...

If you want a load into a new table, then create the empty table AND create the Clustered Index.  Then, forget SSIS... use BULK INSERT with the TABLOCK option and...

@Grant,

I love this article but I'd like to add a nuance to it, especially after what Rick and I were just "talking" about in some of the sayings we listed.

A...

And I'll add a...

Thanks for your solution! Looks awesome. The only one disadvantage is necessity to give names to columns manually. Is there any way to do it automatically? Preferably without dynamic-sql.

[quote-0...

I don't mind changes to software etc. that actually improve the product, but so many changes these days seem to occur just to give the developers something to do. ...

I've personally found that partitioned tables...

I have no real idea except to say that yours isn't the first post where On-prem performance blew the doors off of cloud performance in similar tests.  You might want...

Have you looked to see if the indexes on the table are the same?  Have you updated statistics lately on either?

Heh... apparently, the OP has left the building. 😀

IMHO, take the time to force the MAX to be out of row and default the error to an empty string to prevent the pointer from being "expansive".  Have max...

If you're going to do lots of date-related analysis, I'd just create a Calendar table with all the columns you want to group by, and then join to that...