I am new to the whole idea of dimensions. I have an hour dimension that tells the hour of the day (0-23) a lab test was run. I have a dimension table that basically just had columns HourKey and Hour, where HourKey=Hour. I just added a third column, shift which tells what shift each hour is in. The rational behind the identical columns was just so i had a dimension table. I don't know if that is correct but now i want to use this new shift column in my cube. Do I get rid of my Hour column? How would I setup my heirarchy with the shift attribute? Thank you very much.
PS Its hard to explain why i have a separate hour dimension, i also have a date dimension, trust me on that one
PPS Does anyone have any tips for changing tables and updating them in the DSV and cube without having to readd them and remake dimensions?
You only need a single HourKey column. For any attribute, the name binding is optional, and if unspecified, will reuse the key column. Make Shift another attribute which is related to the Hour attribute. You have to rebuild dimension each time you change its metadata, adding a new attribute is changing metadata.
No comments:
Post a Comment