Backup/recovery strategy would be much easier and reliable with one databse per store.

really?

sumit.joshij (4/2/2010)

---

SO when we create the relation between the PK and FK it check the statistics for PK and FK. it can take time for a huge tranasaction database because...

in short, hire a senior DBA with proven experience in data modeling and physical implementation 🙂

Here are my two cents.

Reading your post I come to the conclusion you have designed a star-schema data warehouse following the dimensional model path. I also assume that having nine...

This is more likely application side.

Check the value of SqlCommand.CommandTimeout property - I think default value is about 30 secs.

A value of zero means unlimited.

You can alter SqlCommand.CommandTimeout value, test...

I have seen people following that path when building a test environment where they are certain about the quality of the data.

Rationale behind is that building PKs/FKs afterwards - with...

The general guidance on partitioning is that you partition a table either to help during quering or to help during archiving/purging... or both.

In general a date column is a good...

I would start by checking requirements with business.

If business is Okay with getting reports on "yesterday's" data depending on database size you can just restore a full backup on "reporting...

Exactly - the combination of the three flags a.k.a. attributes, defines the status of the FK.

The standard setup for FKs between FACTtual and DIMensional tables in data warehouse environments is...

Paul White NZ (4/1/2010)

---

PaulB-TheOneAndOnly (4/1/2010)

---

Actually - in the Oracle world, you have to define such particular FK...

Just out of curiosity... any particular reason you foresee the need of renaming REALDATABASE in the future?

David has my vote on this one.

easy money... nobody uses 11i 'cause it doesn't exist 😀

Actually - in the Oracle world, you have to define such particular FK as "rely" for this...