There are a couple of ways to do that. How big are the databases? If not too big, then backup and restore the production database onto the development database.

Why would it be difficult? You can initially create the partitions, specifying more than you need.

Then run this script regularly:-

SELECT

t.name AS TableName, i.name AS IndexName, p.partition_number, p.partition_id,

i.data_space_id,...

Here's what worked for me:-

CREATE TABLE [dbo].[XML_Convert](

[Kachunk] [xml] NULL

) ON [PRIMARY]

GO

DECLARE @Text XML

SET @Text =

'<SEARCHREQUESTRESULT><DOCUMENTS>

<DOCUMENTDETAIL>

<DOCUMENTDATE/>

<CONSIDERATIONAMOUNT>0</CONSIDERATIONAMOUNT>

<BOOK/>

...

I wouldn't recommend creating partitions when they are needed. Would your query have to wait until the partition is created until it could insert its record?

It would be better to...

I'd implement compression first, as it will reduce the number of reads that you have to do when reading data from database A (on the old collation) to database B...

Cool, I really wasn't looking forward to the Data Warehouse exam!

OK, make sure that you are deleting random rows (not all in one block).

Also insert randomly as well.

Failing that, you could create a new table with a GUID as a...

Ah, I see. How about:-

SELECT C1.ID, C1.[Resource]

FROM dbo.[Consumer] C1

WHERE NOT EXISTS

(SELECT ID

FROM dbo.[Consumer] C2

WHERE C2.[Resource] NOT IN ('MI','RG')

AND C2.ID = C1.ID)

AND...

Can you post the script you are using?

You can add partitioning to a mirrored database as normal, all DDL changes will be pushed across.

This is also true for adding files & filegroups. Just make sure that the...

How about:-

SELECT

ID

FROM

dbo.[Consumer]

WHERE

Resource in ('RG','MI')

Have you tried deleting a large set and then inserting more records (with random PK values)?

How big are the tables?

Then you can shrink the database, this will severely fragment all the indexes.

Based on what you said:-

SELECT

ID

FROM

dbo.[Consumer]

WHERE

ID = 119319004

AND

Resource = 'RG'