Full proc - there is a dependency on a monitoring table, but just running this proc defn will create a hanging tran in build 4305 that doesn't happen in 4000...

Given that you say you are limited pretty much to staying home, then it really boils down to an area you would like to study.

I'll leave aside musical endeavours, because...

Bad plans vs good plans. Knowing the difference and how to transform one into the other. (Includes re-writing monster queries into efficient multi-step processees with temp tables, table vars, etc,...

Just for laughs, I include the test code I used to demonstrate the two (related) points for the NORECOMPUTE situation. 1) that stats going out of date don't force a...

OK, Thanks for the info. I thought that was probably the most likely behaviour but I wasn't entirely confident there wasn't an outside chance that the optimizer decides to ignore...

Thanks for the reply. I totally agree about the dangers of using NOLOCK in any application code. However as a tool for an admin doing a quick check to see...

This works:

declare @prog_name nvarchar(500), @job_id varbinary(32);

select --*

top 1 @prog_name = program_name

from sysprocesses

where program_name like 'SQLAgent - TSQL JobStep %';

--OR

--set @prog_name = 'SQLAgent - TSQL JobStep (Job...

diddybase - < 1Gb, especially the "test" or "proof of concept" databases, beloved of developers, with less than 10,000 records in each table, or with an entire dataset < available...

Just a final point. My initial response to matt stockholm's suggestion above - i.e. that using the $PARTITION fn in the where clause didn't help, was actually wrong. I must...

OK, so I found a relevant page on the Microsoft feedback forum at: Feedback: Partition Table using min/max functions and Top N - Index selection and performance.

Edited highlights:

Description: Partitioned Tables...

Well, to be honest, given that I need to get a workaround at the moment (it being Friday afternoon an' all), I decided to go for a "how long is...

ALZDBA (6/5/2009)

---

If you created your new non-clustering index without specifying a filegroup, you should see a parallel execution for that query, because will actually put this index on a per...

sorry, I'll try that attaching the plan bit again.

So I created a non-clustered, non-unique index on the ID column, like so

USE [Exchange]

GO

CREATE NONCLUSTERED INDEX [IX_GBEOrderNotificationID] ON [dbo].[GBEOrderNotification]

(

[GBEOrderNotificationId] ASC

)WITH (STATISTICS_NORECOMPUTE = OFF

, SORT_IN_TEMPDB = OFF

, IGNORE_DUP_KEY =...

OK, well I have the plans. Not entirely sure how to include them.

The base queries:

declare @topid bigint

print getdate()

select @topid = isnull(max(NotificationSequenceID),0)

from Exchange.dbo.NotificationSequence

print cast(getdate() as varchar) + ' Max NotificationSequenceID:...