Unnecessary persistence is an anti-pattern. Taking this to its extreme is called "orthogonal persistence" - everything persists, including all the finite state machines. This raises the question of how to find a consistent cut.
Sometimes it's called transparent persistence, or persistence ignorance.
The wikipedia article on persistence [] has a section on orthogonal persistence and claims:
The advantage of orthogonal persistence environments is simpler and less error-prone programs
Unfortunately there is no mention of all the serious, inherent problems with orthogonal persistence. For example, it almost certainly involves the Inadequate persistence system anti-pattern.
Nevertheless there's an enormous amount of literature and products trying to achieve it:
The following quote is by Robert Di Falco
I've never seen a transparent Persistence Mechanism that performs adequately - unless that is, it was hand coded to the domain it was serving. The problem with transparent persistence is that you usually lose all the cool things about relational calculus that makes data queries fast - out left joins, searching, and such. The worst thing about transparent persistence is the traffic between the data engine and the application model. For example, you either end up retrieving objects one row at a time or retrieving the entire set of objects - when you are dealing with millions of records, that is a drag.
There have even attempts at making an OS support it: