Multiple fact types in one fact table
2 posters
Page 1 of 1
Multiple fact types in one fact table
Hi,
I'd like to get guidance on the best way to set-up a datawarehouse, with 3 different fact types.
These fact types have 95% of their fields as the same, ie product, customer etc.
To build the reports, I also need to link 2 fact types at a time.
Question is: should i load these 3 diff fact types to one table or have each fact table have it's own table.
What are the pros and cons?
Am thinking more of having separate fact tables - as i will be joining data later on within the same table. ie. sales and inventory joins.
Thanks!
Maureen
I'd like to get guidance on the best way to set-up a datawarehouse, with 3 different fact types.
These fact types have 95% of their fields as the same, ie product, customer etc.
To build the reports, I also need to link 2 fact types at a time.
Question is: should i load these 3 diff fact types to one table or have each fact table have it's own table.
What are the pros and cons?
Am thinking more of having separate fact tables - as i will be joining data later on within the same table. ie. sales and inventory joins.
Thanks!
Maureen
MaureenSy- Posts : 2
Join date : 2013-02-18
Re: Multiple fact types in one fact table
Generally it is a really bad idea to try to put disparate information in the same fact table. Especially if you are talking about things like sales and inventory. Keep them separate.
Re: Multiple fact types in one fact table
Thanks for the response!
Just to make sure i can rationalize the decision, given that they will be joined to produce derived measures. (so business might say the data is not exactly unrelated)
1. if i extend the column to hold the additional measures for the same set of products - this will complicate loading process with respect to joins for column population
2. if i extend the rows, to have several fact types within the same table - this will complicate the report creation due to join within the same table and impact performance
Did i get the Cons correctly? are there other keys considerations?
Just to make sure i can rationalize the decision, given that they will be joined to produce derived measures. (so business might say the data is not exactly unrelated)
1. if i extend the column to hold the additional measures for the same set of products - this will complicate loading process with respect to joins for column population
2. if i extend the rows, to have several fact types within the same table - this will complicate the report creation due to join within the same table and impact performance
Did i get the Cons correctly? are there other keys considerations?
MaureenSy- Posts : 2
Join date : 2013-02-18
Re: Multiple fact types in one fact table
Yes, those are cons. I am sure there are a whole lot more. The thing is, there are no 'pros' for doing it.
Similar topics
» 'Routing' the grain of the fact table to multpile members of multiple dimensions causes the fact table to 'explode'
» How to handle multiple aggregations for multiple KPIs in fact table
» Single fact table vs multiple fact tables - what is the right way in a dimensional model?
» Multiple Fact Tables vs. Consolidated Fact Table
» One Fact table - or multiple?
» How to handle multiple aggregations for multiple KPIs in fact table
» Single fact table vs multiple fact tables - what is the right way in a dimensional model?
» Multiple Fact Tables vs. Consolidated Fact Table
» One Fact table - or multiple?
Page 1 of 1
Permissions in this forum:
You cannot reply to topics in this forum