Using Drupal to pull existing database info?

I mean, will the daily additions to the back-end data be additions, or revisions to existing rows?

If it really is just additions of new data, I'd seriously consider just working with the Node Import module (http://drupal.org/project/node_import), and then try to automate that daily routine.

It's definitely possible to have Drupal pull in external data -- it's the approach I started with for this site, which draws on data for 14,000+ school districts. (See http://drupal.org/node/18429 for details on connecting to multiple DBs.)

The problem is that, when your data isn't actually in Drupal, there's no easy way for Drupal to know what to pull in and when. WIth the education site above, we created nodes that just contained each district's name, state, and then a unique school-district ID that was used in the external DB tables as well. That gave us enough so that Drupal could provide the state-by-state browsing, searching by name, etc., and then we used custom calls in the node.tpl.php files to key off that district ID and pull in all the other data.

My thinking was similar to yours -- the other data didn't change that often, and when it did, it was across hundreds or thousands of districts, so we would never want to edit node-by-node in Drupal. But then when we wanted to search by something other than name, rank or compare districts, etc., there was no way to do so. I kicked around the idea of a "bridge module" for Views with webchick and others, but the consensus was that it would be an insanely complicated undertaking.

So ultimately we created gajillion-field CCK nodes, and imported all the data into Drupal proper. And on those rare occasions when we need to update existing data in bulk, I just do it directly at the DB level in phpmyadmin.

That's not an actual answer, I know. But hopefully it gives you some ideas, and helps explain why no easy "use this module!" answers have been forthcoming.

Good luck...