Keeping data in memory instead of a database
A blog by Martin Fowler about how memory has become an alternative to a database.
It means keeping data in memory instead of storing it in a database and paying an overhead (both in performance and productivity) for reading it, as well as mapping it to objects. Instead, just the log of changes is persistent, using Event Sourcing.
(Rant: I was about to blog about this and the bastard stole it from me... ;-))