yes and no.

• adding a default constraint to an existing column doesn't change anything
• adding a new column and a DEFAULT CONSTRAINT with a single statement, doesn't change the pages...

• Dev-Server with MaxDOP = 8 (of 32 logical CPUs)
• 768 GB RAM
• SQL 2025 Developer (Enterprise), always most current CU
• Simple Recovery for the database
• Log-File on SSDs, the...

• This reply was modified 1 weeks, 1 days ago by Thomas Franz.

• on my test server the query with the fn_Tally function runs 4:30 min
• using

  DROP TABLE IF EXISTS dbo.BigTable
  DECLARE  @StartDT   DATETIME = '2010' ...

Fun fact: Auto close can cause your monitoring software to throw several of false alerts when the database is closed after the app closed its connection.

Took me a while several...

One of main problems of the SQL code parser is, that it doesn't save intermediate results for columns that are calculated per CROSS APPLY (or in a CTE or a...

• This reply was modified 1 months ago by Thomas Franz.

Besides using it to call functions (e.g. CROSS APPLY GENERATE_SERIES() or STRING_SPLIT() or your own ones), I often use it to calculate intermediate results instead of repeating the code multiple...

A few days ago I explained the concept of an untrusted foreign key constraint to a sales person from a vendor:

All doctors in a hospital should hold a doctorate. However,...

for the sake of completeness:

SELECT BIT_COUNT(ASCII('A'))

will return 2 since ASCII('A') = 65 = 100 0001 = 2 bits set

Disabling an index is the same as dropping it, except that it's meta data are preserved, so that you can rebuild it later without the need to know the exact...

To be honest I had choosen Matrix, when it would have been in the list.

Didn't someone (the Oracle or the "god") said something similar regarding Agent Smith in the 3rd...

For the sake of completeness:

To get the original string, you need to use

SELECT CONVERT(VARCHAR(100), BASE64_DECODE(@encoded))

which will return 'Hello SQL Server 2025!'

It's nice to have such a tool in the toolbox (just in case), but I find it hard to find a real use case.

Maybe to compare public names as for...

From the conclusion in the original article:

Keep in mind to delete all existing target backup files before you launch the next backup statement or backup job using a new compression...

Use the OUTPUT Inserted.* INTO #tmp on an INSERT when you need the IDENTITY / NEW_ID / SEQUENCE values for the new records (e.g. for further processing / creating referencing...

Before you all add now 16 other columns to your PK, please remind that you still must not have more than 900 bytes in a clustered index 🙂

Clustered Index:

Msg 1944,...