In this paper, we investigate efficiently maintaining multiple versions of B+-trees.We propose two main schemes: the fat field method and the pure version b~ock method.The former method is suitable for range queries or whenever lookup on a released version is an important issue.The latter method is more space efficient, and is suitable when the total number of keys from all the versions are small.

A knowledge-base management system (KBMS) with the following goals is proposed: (1) to allow a knowledge engineer to update a knowledge base and have these updates persist on secondary storage, (2) to allow multiple knowledge engineers to have shared access to a knowledge base and modify the knowledge base concurrently, and (3) to maintain consistency of the shared knowledge base as it evolves. At the heart of the KBMS is the version store, which is a persistent storage structure that maintains multiple versions of a knowledge base. Retrieve and update operations are defined on the version store to efficiently access and modify any version. Objects in the version store are clustered to support efficient access of an entire version or some subparts of it. To store the effects of an update requires space proportional to the size of the update and which is independent of the size of the knowledge base. The additional cost of maintaining multiple versions is within a small constant factor of maintaining a single version.>

We examine several behaviors for query systems that become possible with the ability to represent and reason about change in data bases: queries about possible futures, queries about alternative histories, and offers of monitors as responses to queries. A modal temporal logic is developed for this purpose. A completion axiom for history is given and modelling strategies are given by example.