Period/Time Dimension Grain

Hi everyone!

I am in a situation with fact tables that have grain at different time period levels i. e. some daily, others monthly and some quarterly. Therefore, I am trying to determine if it is OK to use a single period dimension (with the date level grain) with all these facts? Or is it worth to have different period dimensions i. e. one for date level, one for month level, one for quarter level?

Please let me know how do you model this in your data warehouses. My current client wants to have a single period dimension (at the date level) and have it "role-play" with the facts that have grain at the higher level time periods. I'd like to recommend them an option with multiple time period dimensions, daily, weekly monthly, depending on the fact grain required ...

Essentially I am working with the forecasting data. Forecasting is done at the academic session level, not dates ... But the client has a single date dimension with all the possible periods including the academic session. However this dimension grain level is a day (or a date).

Please feel free to let me know your thoughts or share your experiences.

Thank you,
Mike

Just realized that pretty much the same question was already asked/answered here.
http://forum.kimballgroup.com/t1572-using-the-date-dimension-for-summary-fact-tables-or-new-specialized-month-dimension

Looks like it better to have a single date dimension role-play as higher level period dimensions.

Sorry about the "duplicate" post everyone.

Welcome to the beauty of dimensional modeling where detail facts aggregate seamlessly.