If you know the dates are always in DD/MM/YYYY format, perhaps you could pre-swap the month and day segments of the string & avoid the need to set dateformat --...

If you know the dates are always in DD/MM/YYYY format, perhaps you could pre-swap the month and day segments of the string -- e.g.,

DECLARE @DateString CHAR(10) =...

See https://www.sqlservercentral.com/articles/cross-tabs-and-pivots-part-1-converting-rows-to-columns-1 & https://www.sqlservercentral.com/articles/cross-tabs-and-pivots-part-2-dynamic-cross-tabs

Seems related to your other question: https://www.sqlservercentral.com/forums/topic/assigned-as-guest-when-installed-sql-server-on-workstation.

If his windows login has sysadmin permissions, then yes. If not, then somebody (e.g., IT) has to login as sa to create users &...

frederico_fonseca wrote:

do not use the native string_split - slow and does not return the ordinal of the strings.

Aaron Bertrand's tests from 2016 showed exceptionally good performance: https://sqlperformance.com/2016/03/sql-server-2016/string-split

Might have been appropriate for Halloween.

Always exclude SQL Server data, backup, & repl folders from antivirus scans -- they are well known to be devastating to performance. Many people have experienced this -- it's just...

You don't know the delimiter? It's not predefined/standardized?

STRING_SPLIT function is available in 2016, but would require nvarchar instead of ntext.

ntext has been deprecated since SQL Server 2008, & has limitations...

Good question, homebrew01.

roland:

Recovery model? Full? Simple? Bulk-logged?

I assume we're talking about full backups. How often?

If full, how often are you doing log backups? Are they succeeding?

Is backup the only thing happening at that time? Is there index maintenance or CHECKDB going on in the same job (or worse, simultaneously)?

What do you mean by "own backup...

A non-clustered index may help.

In most cases, you should probably have a clustered index.

SSMS can tell you dependencies for the procedure as a whole.

At the CTE level, as far as I know, that will require a diligent data engineer.

Be alert to the use...

Duplicate of https://www.sqlservercentral.com/forums/topic/refinement-of-my-query-in-a-best-way-with-optimum-results-for-large-data

Is your data really so bad that all those predicates could be null or empty strings? Do they really contain invalid commas?

All the ISNULL(), REPLACE(), & LOWER() functions make your...

• This reply was modified 2 years, 10 months ago by ratbak.

Note that query would still return a row for every different combination of the columns. If all you're trying to get is the sum, you need to change the group...