Implementation of Raw Data Vault

All, I am new to DV and if we have to design RAW data vault, what that would be? sharing the logical erwin model

Here are the assumptions being made so far:
Tender as Hub?
Site as Hub
Store is child or supertype of Site
Region not sure, need help since this has flatten data , please see below

Transaction not sure, need help
CustomerTransactionFile or CTF is a reference id that comes in our transactional data files - it is generated at the store and is used to tie the transaction records in the 28 files together.

@patrickcuba or other DataVaulter looking for some guidance on this. Thank you in advance.

looks like ref data — do you really want to vault this?

If not, are you saying we should put all the reference tables to a schema and lookup and directly join them instead of vaulting it.

Also, could you please shed light on other questions as well.


Despite @patrickcuba already answered, would like to hint you in basics:

  • for a link to exist there must exist at least one unit of work/transaction/relation between 2 hubs, if you just read (discard er diagram) you say store emits files with some id generated by store good, but then theres no 2nd hub or destination of those files, therefore theres not enough hubs to introduce a link, therefore you just have 1 hub with 1 satellite that keeps generated files, and sat using several reference data.
1 Like