Thanks David. I’ve thought about the hierarchy you mentioned but the hierarchy can change and I need to capture it as it was when the data was generated.
I should have pointed out in my last response that I was wondering if the performance of the pk index on work_session would be better if my primary key was (customer_id, work_session_id) or if (work_session_id, customer_id) will be fine. Customer_id will be repeated quite a bit in the table but work_session_id should be unique across the whole table. Thanks, Craig From: David G. Johnston [mailto:david.g.johns...@gmail.com] Sent: Monday, August 8, 2016 12:43 PM To: Craig Boucher <cr...@wesvic.com> Cc: pgsql-general@postgresql.org Subject: Re: [GENERAL] Column order in multi column primary key On Mon, Aug 8, 2016 at 3:06 PM, Craig Boucher <cr...@wesvic.com <mailto:cr...@wesvic.com> > wrote: Here is one of the tables that can have millions of rows and foreign key constraints to 5 other tables. CREATE TABLE public.work_session ( work_session_id integer NOT NULL DEFAULT nextval('worksession_worksessionid_seq'::regclass), customer_id integer NOT NULL, store_id integer NOT NULL, department_id integer NOT NULL, station_id integer NOT NULL, CREATE TABLE public.department ( department_id integer NOT NULL DEFAULT nextval('department_departmentid_seq'::regclass), customer_id integer NOT NULL, No low-hanging fruit that I can see. This is just some thought food - much would depend on the usage patterns of the table and the exact relationship between the different attributes. Specifically, you've modeled this as an intersection of 4 orthogonal aspects (customer, store, department, station) while in reality these items likely exist in a strict hierarchy. It would be possible to have the only FK on work_session be (station_id, customer_id) - and have station have the FK (department_id, customer_id) - and so forth. David J.
