I have an application that makes heavy use of foreign keys all over
the tables. This is very nice since the data is very consistent. There
also this "central" table which holds "sites" in it. A site pretty
much is the crux of it all. Deleting a site will very precisely
eliminate all data regarding it, since there's CASCADE on delete's
everywhere.
The only trouble I'm having is that the original developers apparently
didn't account for large amounts of data. I'm starting to get a LOT of
data in some tables, and nowadays deleting a site will take a
disgusting amount of time (in the range of tens of minutes). It's
impossible to do it via Web, so I have to issue the central delete
from the shell and leave it running until it's done.
Is there any way I can make things better? I could queue site drops
and have a cronjob pick them up instead of deleting "live" via Web,
but that's just silly patchwork IMHO.