[RBASE-L] - Re: Schema Design Question: Look-ups

[Doug Hamilton]

Like I said, I haven't tried this, but just going through it as a thought experiment...and assuming it's a network environment... You would read a master table from the server once and create a local lookup table(s). After that, there's only the local user whacking away at the table(s). And no network traffic resulting from subsequent look ups - that's good for all users.   'Course, a fast enough network may rival local disk speed Judicious use of indices on the master table will help extract data quickly when creating the local temp table. Ditto those indices when looking up in the local table. Drawback: If the lookup data is updated "often" (more than once a day?) and other users are dependent on those updates, you'd have to figure out a way to push those updates to the other users.  But then maybe that data is better in one central lookup table.  A ZIP code table would probably fall into this category. A blanket statement about the integrity - whatever you can do on a permanent table you can do on a temp table. Have a great Labor Day weekend! Doug Bruce Chitiea wrote: Larry, Emmit, Dennis, Karen, Jim, Doug: Thanks much for your wisdom on this. My "self-tutorial" schema is already two-thirds "z" (eg: zcounty, zstate, zzipcode) lookup tables, and I'm wondering at the coding load required to maintain and access those tables. Any which way, there's work to be done. >From what you say, there seems to be a workable middle where integrity, flexibility and performance may just possibly meet. Larry: Do Karen and Doug's approaches adequately address the integrity issue? Doug: Do you see a performance issue? My weekend is now complete! Yours, bruce chitiea safesectors inc. -------- Original Message -------- Subject: [RBASE-L] - Re: Schema Design Question: Look-ups From: Doug Hamilton <bugl...@wi.rr.com> Date: Fri, September 03, 2010 3:21 pm To: rbase-l@rbase.com (RBASE-L Mailing List) Why not use a Master table ...