Question

I fear I don’t know what I’m doing.

1:

I have a table called ticket which has a column called total. When the total is updated I want to keep a record of it (old total, etc), so I decided to remove the total column and create a table called ticket_total with columns ticket_id, total, and datetime (the most recent of course is the "current" total).

or

2:

Then I realized that I will later want to give my clients the ability to sort tickets by total, or pull reports that aggregate the totals, etc. So, I decided instead to put back the total column on ticket, and to change the total column directly when the total is updated, but first create a ticket_total row as a record of the previous total.

It seems that version 2 would be highly efficient because I wouldn’t need to query the related ticket_total table as much, but I wonder what you DB gurus out there think. I’m just learning database design and fear I’m never going to be good at it.

Answer 1

I would go with option 2 that you have suggested.

Just make sure that you are doing the Update (of ticket) + Insert ( in ticket_total ) in a transaction to ensure that the integrity is maintained.