Forum Discussion
why dimension table
Hi Anonymous,
Each dimension should relate to a specific Domain, eg all customer attributes would belong in the Customer Dimension, all Vendor attributes would belong in the Vendor Dimension, all date attributes would belong in the Date Dimensions .... Each Dimension would then join back to the fact on the unique key per dimension member. If you find that you are creating many single attribute dimensions you should try to find logical groups and derive a dimension key from the unique combinations.
Hope that Helps,
Did I answer your question? Mark my post as a solution!
Did my answers help arrive at a solution? Give it a kudos by clicking the Thumbs Up!
Many thanks! So I understand if I need to use multiple filters from the same table, the best practice is to create seperate dimension tables.
the next question is how many dimension tables I should create? I have seen people create lots of dimenstion tables, each table corresponds to one filter. The problem is that will generate a complicated relationship schema. The other solution, I guess is to create only a few dimension table (e.g., put country, job code, company name in one reference table), however, the reference table should contain all the logic combinations of country-job-company. Is that correct?
Hi Anonymous,
Each dimension should relate to a specific Domain, eg all customer attributes would belong in the Customer Dimension, all Vendor attributes would belong in the Vendor Dimension, all date attributes would belong in the Date Dimensions .... Each Dimension would then join back to the fact on the unique key per dimension member. If you find that you are creating many single attribute dimensions you should try to find logical groups and derive a dimension key from the unique combinations.
Hope that Helps,
Did I answer your question? Mark my post as a solution!
Did my answers help arrive at a solution? Give it a kudos by clicking the Thumbs Up!