Future information processing environments will consist of a vast network of heterogeneous, autonomous, and distributed computing resources, including computers (from mainframe to personal), information-intensive applications, and data (files and databases). A key challenge in this environment is providing capabilities for combining this varied collection of resources into an integrated distributed system, allowing resources to be flexibly combined, and their activities coordinated, to address challenging new information processing requirements. In this paper, we describe the concept of distributed object management, and identify its role in the development of these open, interoperable systems. We identify the key aspects of system architectures supporting distributed object management, and describe specific elements of a distributed object management system being developed at GTE Laboratories.

A mechanism for producing views in an object-oriented system is presented. The results are analogous to database views in traditional database systems, except that the object views presented hide or expose methods as well as data. The mechanism is based on the object model's built-in facilities for defining data and procedural abstractions and for constructing new types and objects. It uses the type system and the query language of the model to support arbitrary transformations of the underlying representations in defining database views, without introducing any new mechanism specifically to support views. Careful use of the query language allows one to define updatable views. Also indicated are how abstraction and view mapping capabilities can be used to support federation of heterogeneous software and databases.>

An object-oriented approach to management of engineering design data requires object persistence, object-specific rules for concurrency control and recovery, views, complex objects and derived data, and specialized treatment of operations, constraints, relationships and type descriptions. We discuss object-orientation as more than an implementation paradigm, and show how an object-oriented approach simplifies both use and implementation of engineering design systems.

Part Hierarchies are a fundamental datatype in CAD applications. But intelligent and efficient processing requires major extensions to DBMS data models, query languages, and processing algorithms. We explore formulations and execution algorithms for path-traversal queries. Hierarchy semantics are then exploited for spatial data and to intelligently choose an appropriate detail level for query output.

Many capabilities that are needed for recursive applications in engineering and project management are not well supported by the usual formulations of recursion. We identify a class of recursions called “traversal recursions” (which model traversals of a directed graph) that have two important properties they can supply the necessary capabilities and efficient processing algorithms have been defined for them. First we present a taxonomy of traversal recursions based on properties of the recursion on graph structure and on unusual types of metadata. This taxonomy is exploited to identify solvable recursions and to select an execution algorithm. We show how graph traversal can sometimes outperform the more general iteration algorithm. Finally we show how a conventional query optimizer architecture can be extended to handle recursive queries and views.