Use an AFTER trigger(s) for both INSERT and UPDATE.

For example:

--USE [<your_db_name_here>] --naturally make sure you are in the right db

SET DEADLOCK_PRIORITY LOW --probably irrelevant, but just in case

DECLARE @list_missing_indexes bit

DECLARE @table_name_pattern sysname

--NOTE: showing missing indexes can take some...

You will need to look at least at the stats SQL provides for indexes:

1) missing index stats and

2) index usage stats.

Between those and you knowledge of the table, we...

Lowell (4/16/2015)

---

Talib123 (4/16/2015)

---

Not me 3rd Party. The table is a Heap it is badly designed with far too many columns and serving many different purposes.

They believe it will improve...

Someone's apparently discovered an obscure bug that's being repeated here. Software bugs are just bugs, we can't "explain" them.

There are many, many times when the PK should not be the clustered index. Why do you think it should be changed??

You're welcome. The really great thing is that that code still works perfectly if/when you convert the column itself to datetime (or date) ... nice!

You want to use the current index where possible, so do this:

select MRN, Name, AppointmentDate

from DATA

where datetime >= '2015-04-01 00:00:00.000'

and datetime < '2015-05-01 00:00:00.000'

Detach now (since SQL 2005, IIRC) changes the security on the files -- this is an intentional thing by MS, as a "security feature". But it's been a royal...

You check for a "GLOBAL" cursor but you don't explicitly specify a GLOBAL cursor when you DECLARE it. Whether a cursor is local or global can vary based on...

You should look into RCSI as well. That may solve your issue even more completely and with less effort during coding :-).

CREATE FUNCTION [dbo].[CheckLoginDetails_Username] (

@SuppliedUsername nvarchar(150),

@SuppliedPassword nvarchar(150)

)

RETURNS bit

AS

BEGIN

RETURN (

SELECT CASE

...

Maybe a join to reduce repetition of comparisons:

SELECT bf.BF_ORGN_CD, bf.BF_BDOB_CD, bf.BF_TM_PERD_CD, bf.data

FROM BF_DATA bf

INNER JOIN (

VALUES('A1', 'B1', 'C1'),

...

There used to be both recoverability and performance reasons for the split. Now it's more about recoverability than performance.

If possible, you want the data and logs on separately "fail-able"...

Eirikur Eiriksson (4/13/2015)

---

ScottPletcher (4/13/2015)

---

Eirikur Eiriksson (4/13/2015)

---

ScottPletcher (4/13/2015)

---

Here's an alternative approach, also using CROSS APPLY, that can take advantage of the (presumed) clustered index on prefix.PREFIX_VALUE; no guarantee, but it might...