Following my previous question here:
Here\'s a screenshot of my billing-enable appengine console:
As far as I know, resident instances are for times when no other instances (f. e. dynamic) are available (busy, no one started). They are the buffer between full utilization and new (dynamic) instances available. If every dynamic instance is at its limit (or no one available), the app engine needs some time to start new instances. In this time the resident instance is getting new requests so that your service is not getting busy for the users (and/or to decrease service response time). A resident instance with low usage will cost you as much as one with heavy usage.