Actually, best is often to force max values out of row, for every row.  That is generally my preference, although naturally there are a few specific exceptions.

Would have to see the query's use of col1 thru col8: in WHERE, JOINs, GROUP BY and even the SELECT itself.

SQL Server's missing index feature doesn't properly "understand" the importance...

% returns a remainder from division.

So, 1.1 % 1 yields 0.1.

2.54 % 1 = 0.54

3.0 % 1 = 0.

Came across an actual query which gave me an idea about another version of the test script for DISTINCT vs. GROUP BY.

create table #BalanceType (
id int...

SELECT j.name AS job_name, jh.*
FROM (
    SELECT jh.job_id, MAX(jh.instance_id) AS instance_id
    FROM msdb.dbo.sysjobhistory jh
    WHERE jh.run_status = 0...

You need more than just year to accurately calculate age.  Some people born in 2011 are 10 years old, but some are only 9 years old.

My code works fine with that format of string:

;WITH test_data AS (
    SELECT * FROM ( VALUES('(b) Joe Brown -12563'), ('(a) Mary Edwards -15425'),...

Make catalog changes to that your new tempdb file set up takes effect the next time SQL starts up.  You just have to delete the old tempdb files yourself.  I...

SQL provides a CASE expression to test for things like that:

SELECT ..., CASE WHEN PERSON.TERMINATION_DATE > '19000101' THEN 'N' ELSE '' END AS TERMINATION, ...

LTRIM(RTRIM(SUBSTRING(name, CHARINDEX(')', name) + 1, PATINDEX('%[+-][0-9]%', name) - (CHARINDEX(')', name) + 1))))

;WITH test_data AS (
    SELECT * FROM ( VALUES('(b) Joe Brown -12563'), ('(a)...

It depends.  If there's a large volume of data written to your SQL log even when there are no errors, then you should probably switch the logs more often (and,...

No directly usable data to test with, but I think this should at least be close:

SELECT PH.*, PA.*
FROM dbo.PolicyHeader PH
OUTER APPLY (
    SELECT TOP...

Add WITH (TABLOCK) hint after the table name.

Examples:

UPDATE dbo.table_name WITH (TABLOCK)

SET ...

INSERT INTO dbo.table_name WITH (TABLOCK)

SELECT ...

I think you can combine the first two queries into a single query, also avoiding having to rank all the totals:

WITH cteCombined AS
(
 SELECT TOP (1) WITH...

For the much more straightforward approach:

SELECT DISTINCT N

vs

SELECT N

...

GROUP BY N

my machine shows a slight edge for DISTINCT.  That's what I've seen in normal usage, too.  For a straight list...