Keynote Discussion on Evolution in Persistent Systems

Abstract

Evolution is without doubt one of the hardest problems to address with respect to programming systems. When the systems under consideration handle large amounts of long-lived data, the ability to handle evolution becomes not only essential, but also even harder to provide. When some of the intrinsic restrictions of traditional databases are removed, as is the case with persistent programming systems, the problems become harder still.

Very little work has been done on system evolution in persistent systems, probably for the reason that it is simply too hard. A major result of this keynote discussion, however, was to show that the current state of thinking about evolution is at least sufficiently advanced for a reasonable level of technical discussion to emerge amongst the workshop participants. Persistent systems research seems to have advanced at least to the stage where problems as hard as these may start to be addressed.