Is distribution a genuine problem for persistent systems or is addressing the real problem?

It is generally assumed that distributed systems are fundamentally different from stand-alone systems, in that they require different operating systems, database systems, etc. This belief arises from experience with conventional systems, where the level of sharing, communication, etc. in a distributed system takes place basically at the file system level. The result is massive complexity and duplication of software functions in most distributed systems. Given an appropriate underlying model, this need not be the case in a persistent object system. Such a model can be based on a large persistent uniform virtual memory. In a stand-alone system this model removes the distinction between a file store and a virtual memory, eliminating much of the duplication which arises in conventional single-node operating systems. Extending this model to encompass distribution involves allowing the persistent object store to be distributed