Research in temporal database management has viewed temporal dynamics from a structural perspective, posing extensions to the entity-relationship (E-R) model to represent the state history of time-dependent attributes and relationships. We argue that temporal dynamics are semantic rather than structural and that the existing constructs in the E-R model are sufficient to represent them. Practitioners have long used E-R models without temporal extensions to design systems with rich support for temporality by modeling both things and events as entities — a practice that is consistent with the original presentation of the E-R model. This approach supports methodologies that leverage narrative and human cognitive processing capabilities in the development and verification of data models. Furthermore it maintains modeling parsimony and facilitates the representation of causality — why a particular state exists.

The allocation of data and operations to nodes in a computer communications network is a critical issue in distributed database design. An efficient distributed database design must trade off performance and cost among retrieval and update activities at the various nodes. It must consider the concurrency control mechanism used as well as capacity constraints at nodes and on links in the network. It must determine where data will be allocated, the degree of data replication, which copy of the data will be used for each retrieval activity, and where operations such as select, project, join, and union will be performed. We develop a comprehensive mathematical modeling approach for this problem. The approach first generates units of data (file fragments) to be allocated from a logical data model representation and a characterization of retrieval and update activities. Retrieval and update activities are then decomposed into relational operations on these fragments. Both fragments and operations on them are then allocated to nodes using a mathematical modeling approach. The mathematical model considers network communication, local processing, and data storage costs. A genetic algorithm is developed to solve this mathematical formulation.>

As database applications have become more sophisticated, the development of support tools for database design has become more critical. Before support tools can be developed, however, the nature of database design problems and their solutions must be well defined. We apply data modelling principles to these tasks. The result is a multi level model of databases. The logical level describes database design problems and the physical level describes their solutions. This model is the basis for a computer aided database design methodology developed by the authors.

In recent years the information processing requirements of business organizations have expanded tremendously. With this expansion, the design of databases to efficiently manage and protect business information has become critical. We analyze the impacts of record segmentation (the assignment of data items to segments defining subfiles), an efficiency-oriented design technique, and of backup and recovery strategies , a data protection technique, on the overall process of database design. A combined record segmentation/backup and recovery procedure is presented and an application of the procedure is discussed. Results in which problem characteristics are varied along three dimensions: update frequencies, available types of access paths, and the predominant type of data retrieval that must be supported by the database, are presented.

Frame memory is a virtual view of secondary storage that can be implemented with reasonable overhead to support database record storage and accessing requirements. Frame memory is designed so that its operating characteristics can be easily manipulated by either designers or design algorithms, while performance effects of such changes can be accurately predicted. Automated design procedures exist to generate and evaluate alternative database designs built upon frame memory, and the existence of these procedures establishes frames as an attractive memory management architecture for future database management systems.

This paper provides an overview of a methodology developed to support systems analysts in the process of database design. The design approach is built upon an analytic model composed of (1) parametric descriptions for components of a generalized database organization, (2) costing equations which can evaluate a proposed modular database design, (3) an analyst interface which accepts an arbitrary database organization for evaluation, and (4) search procedures which automatically generate and compare thousands of alternative designs. Performance is measured as the sum of storage, retrieval, and maintenance costs and is estimated from parameters of the proposed design, the problem description and the storage environment. A virtual, record-frame view of secondary storage has been developed in which data records are added, deleted and modified with minimal effect on existing data structures. Application of the modeling approach to a realistic design problem is described, and modeling accuracy to within four percent is claimed.

It is generally believed that 80 percent of all retrieval from a commercial database is directed at only 20 percent of the stored data items. By partitioning data items into primary and secondary record segments, storing them in physically separate files, and judiciously allocating available buffer space to the two files, it is possible to significantly reduce the average cost of information retrieval from a shared database. An analytic model, based upon knowledge of data item lengths, data access costs, and user retrieval patterns, is developed to assist an analyst with this assignment problem. A computationally tractable design algorithm is presented and results of its application are described.