|
Onamuji
Aged Yak Warrior
504 Posts |
 Posted - 2001-08-02 : 09:28:41
|
| .........................................Solution A (the old way): Budget_Table: - Id, Title, JanBud, FebBud, ..., DecBud, JanAct, FebAct, ..., DecActSolution B (the road in between): Budget_Table: - Id, Title Budget_Table_Entries: - Id, Budget_Table_Id, Type, Jan, Feb, ..., DecSolution C (the new way): Budget_Table_Items: - Id, Title Budget_Table: - Id, Budget_Table_Item_Id, Type Budget_Table_Entries: - Id, Budget_Table_Id, Month, Value..........................................Solution A: This is how the data is currently setup in access.Solution B: This would be ok, but there are 3 different types of monthly values, Budget, Estimate, Actual. The Budget and Actual types have all 12 months, the Estimate only has 4. That's when Solution C came into existance.Solution C: This would support all three types, yet not cause null entries for 8 out of the 12 months for Estimates. All though, this will cause the data to be almost a 3rd bigger, well, for each Budget_Table_Item there will be 3 Budget_Table entries, and for each Budget_Table entry there will be 28 entries in the Budget_Table_Entries table. With my current dataset I have over 11,000 Budget_Table_Items and that makes the Budget_Table have close to 33,000 entries... which will in turn make the Budget_Table_Entries table contain close to 924,000. This concerns me with doing joins to extract the proper data. I think I will be ok, but I don't want to waste a day writing code to import in this form if its not a good idea. As always, thanks for any and all help.Thoughts, comments, questions.Regards,Adam |
|
Design Question (for optimal performance)