Schema Speration vs. Refrential Integrity
36681Sep 28 2004 — edited Sep 28 2004Newbie Question...
I am developing a data model and someone mentioned to me when doing my data model for our specific app to group related objects in their own schema's. For example, Item data and related information like warranty, inventories, etc, should be all on one schema and objects like orders, sales, returns, etc. should all be in another schema and keep relationships within the schema but do not cross relationships over schema's. For example, and Item in on an order, so there would be some sort of integrity. The app should lookup the item number to see if it is valid, and then store the record in the object in the order object through code (pl/sql also). I hope this makes sense :)
When running a query, just simply join the tables in the where clause and to mimic referential integrity. Is this a good practice when developing your database. Is this common? What are some good ways of doing this. Any recommendations? I have about 75 tables in my model and relationships are all over the place, and the person QA'ed my model and told me to do it his way (mentioned above). What should I do???
Please help