Maybe I wasn't very clear. I meant that if you update the normalized warehouse first then populating a mart is as simple as an inner join between those two tables...

Maybe you don't really mean that Paul? Star schema is a design pattern quite commonly seen in relational database design. "Dimensional" models...

First update your normalized, history-preserving view of the data in the Warehouse with the data that changed. Then populate the denormalized star-schema Mart from that. The Mart may even be...

Hi Chris, I suggest you specify periods of time as half-open intervals (also called half-closed intervals). That means x is within the valid time period iff (start_time >= x <...

okbangas (10/20/2011)

---

A conceptual thought: We have now (from SQL Server 2008) got filtered indexes. What do you think of the idea of having filtered foreign keys as well. In the...

jarid.lawson (1/6/2012)

---

I am working on a project, and I've hit a small-ish bump in the road. I have to link a client to multiple product types. I need this to...

David Moutray (12/27/2011)

---

I am trying to convince my managers to allow me to enable READ_COMMITTED_SNAPSHOT on the databases for our production data warehouse. I have two questions for you,...

How would you...

From a SQL developer's perspective it is tempting to assume that data integrity should always be enforced by the database. Unfortunately SQL Server (like most SQL DBMSs) is only really...

Dev (12/25/2011)

---

You can have a Normalized data warehouse but your data warehouse won’t be more than a Data Archival Server. You will kill your BI applications if you...

Dev (12/22/2011)

---

Lamprey13 (12/22/2011)

---

Koen Verbeeck (12/22/2011)

---

<snip>

It is perfectly acceptable to have a denormalized database design in a relational database, for example in a data warehouse. SQL still works, and...

ggpevans (11/19/2011)

---

Hello, I have searched far wide on two topics, with no luck.

The two topics are

a) Relationships

b) Joins (Inner joins and cross joins etc)

I understand that...

You are asking specifically about case-sensitivity for...

I guess it might amuse me too but I didn't notice that...

Not a great start in my opinion:

Relational databases fundamentally have nothing to do with tables being related to...