Primary Key of Dimension and Fact Tables

DavidStein at 21h27
18
Aug
2010
Primary Key of Dimension and Fact Tables

I've seen several people give presentations which purport to represent the Kimball Method and I've seen this done two ways.

What is the current position on what should be the primary key of a dimension table?

I was under the impression that we should have surrogate meaningless integer keys (unique clustered indexes) for each dimension table, but should that value be the primary key as well?

Are Fac
About
This topic belongs to the forum
  • Data Warehousing and Business Intelligence based on dimensional modeling and the Kimball Lifecycle.

  • Numbers of topics : 1842
  • Numbers of messages : 8656
  • Numbers of users : 3787
  • Numbers of points : 1497
Similar topics
Hi My data warehouse will have CONTRIBUTIONS fact table which is sourced from a OLTP General Ledger Table. All transactions from the GL table will be put in a ODS staging table (GL_Stage). The GL table primary key consists of 9 fields (transaction
Can the relationship between a dimension table and fact table be identifying as well as non-identfying?
When designing a fact table, is it best practice to make a composite primary key out of all the foreign keys? Thanks, Kurt V
I am having trouble joining two fact tables. I have read the different topics and see that the bridge table would be my logical option. The subject I have is Orders. I can place an Order which may contain many items...as well as, the items can be a par
The title question is kind of rhetorical because I didn't actually think it was a common practice but I was hoping someone could offer an alternative . Let's say I have a fact_service_provided table where each row describes a service that the business
Forums from same category
See also
more_less
Informations

10 Replies For the topic :
"Primary Key of Dimension and Fact Tables"

This topic has been viewed 2938 times.

Last message :
18/08/2010 at 21h27 by "DavidStein"