Thanks again Martin for the prompt response. So I've read through a lot of the notes around used memory on the support portal, and there's some excellent documentation on the subject. As for the bad sql script procedure, it was easily corrected but i actually kept the bad version around for testing and help with future support.
And yes, our dev box had a very large value for late materialization threshold per note 1975448 (trying to check why/who changed it).
However..
- This example doesn't involve translation tables, I can see these easily in M_DEV_JOIN_TRANSLATION_TABLES_ (Aside: Is there anyway to automatically purge?, table re-distribution seems to work, but it would be nice not to have to do that each time)
- Is there any scripts that can give us more information about these objects and anyway these can also be manually purged, especially in the case where we had a mem allocation error and this large amount of used memory hadn't been returned to the pool?
