You certainly don't want to just store the entire name in one long, indistinguishable string. Trying to accurately parse that out later is virtually impossible.

As to the different rules/formats,...

1) Don't use ISNULL, you don't need it there anyway:

select c1,c2,c3 From numbers

where c3 <> 'CP-961080'

2) A nonclustered index lookup would be very inefficient here, although you might be able...

I believe that SQL itself has an internal limit on how large of a log file it can use, so the max log size is set to that. No...

Probably the easiest way is to use derived tables within a main query, like below, assuming that each derive table returns only a single row:

SELECT q1.*, q2.*, q3.* --, ...

FROM...

In theory, it's also possible that you could get any 1,000 rows that are already present in buffers and thus don't require physical I/O. Without an ORDER BY, any...

Some things to check, roughly in order:

1) Verify that a valid and reasonable "max memory" value is set in SQL Server.

2) Verify that IFI is turned on in Windows to...

I can't imagine any tool that would be able to do that, given that each view level could be arbitrarily complex.

mike 57299 (7/25/2016)

---

I use the index for other queries as well. Can I create a new filtered index? How do I make the function use the new index?...

I agree overall with Alan's comments. If you're willing to convert existing code/queries, also look into an itvf to do the computation.

However, for now, let's tune what you have....

Jeff Moden (7/21/2016)

---

ScottPletcher (7/21/2016)

---

Jeff Moden (7/20/2016)

---

ScottPletcher (7/20/2016)

---

If the "isactive" flag is significant, i.e. a significant number of rows are actually inactive, then cluster first on it. If not, get...

Jeff Moden (7/20/2016)

---

ScottPletcher (7/20/2016)

---

If the "isactive" flag is significant, i.e. a significant number of rows are actually inactive, then cluster first on it. If not, get rid of it....

If the "isactive" flag is significant, i.e. a significant number of rows are actually inactive, then cluster first on it. If not, get rid of it. At the...

You could use a standard unique constraint if you continually deleted searches 2 hours or more old from the table. For example, have a job that deletes rows every...

The "standard/best-practice" way to get yesterday's date is:

DATEADD(DAY, DATEDIFF(DAY, 0, GETDATE()) - 1, 0)

Based on that, then make the necessary 6-hour adjustment on the starting time [the "+" in the...

While logged in as sysadmin, create a stored proc that includes "WITH EXECUTE AS OWNER". In that proc, have the command:

EXEC msdb..sp_start_job @job_name = @job_name, ...

Then GRANT the role(s)/user(s)...