Database triggers / referential integrity and in-memory caching
- by Ran Biron
Do you see database triggers / referential integrity rules being used in a way that changes actual data in the database (changing row w in table x causes a change in row y in table z)?
If yes, How does this tie-in with the increasing popularity of in-memory caching (memcache and friends)? After all, these actions occur inside the database but the caching system must be aware of them in order to reflect to correct state (or at least invalidate the possibly changed state). I find it hard to believe that callbacks are implemented for such cases.
Does anyone have real-world experience with such a setup / real-world experience with considering such a setup and abandoning it (which way did you go? if caching, how do you enforce integrity?)