David K. Hsiao

Federated databases and systems: part I --- a tutorial on their data sharing

The issues and solutions for the interoperability of a class of heterogeneous databases and their database systems are expounded in two parts. Part I presents the data-sharing issues in federated databases and systems. Part II, which will appear in a future issue, explores resource-consolidation issues.Interoperability in this context refers to data sharing among heterogeneous databases, and to resource consolidation of computer hardware, system software, and support personnel.Resource consolidation requires the presence of a database system architecture which supports the heterogeneous system software, thereby eliminating the need for various computer hardware and support personnel. The class of heterogeneous databases and database systems expounded herein is termedfederated, meaning that they are joined in order to meet certain organizational requirements and because they require their respective application specificities, integrity constraints, and security requirements to be upheld. Federated databases and systems are new. While there are no technological solutions, there has been considerable research towards their development. This tutorial is aimed at exposing the need for such solutions. A taxonomy is introduced in our review of existing research undertakings and exploratory developments. With this taxonomy, we contrast and compare various approaches to federating databases and systems.

Heterogeneous databases: proliferations, issues, and solutions

The authors characterize the notion of heterogeneous databases as the inevitable consequence of replacing the traditional data processing practice with modern database management. They articulate the current problems and future issues connected with the great proliferation and overwhelming use of heterogeneous databases and their DBMS (database management systems). They survey available DBMS solutions to these problems and issues. The limitations and advantages of various DBMS solutions are examined. To this end, a taxonomy of DBMS solutions to the problems and issues of heterogeneous databases is presented. Some of the current research and proposals on the heterogeneous-database solutions are discussed in the context of the taxonomy. >

Towards a better understanding of data models through the multilingual database system

An approach to the design of a database system, the multilingual database system (MLDS), has been proposed and implemented. MLDS is a single database system that can execute may transactions written respectively in different data languages and support many databases structured correspondingly in various data models, i.e. DL/I transactions on hierarchical databases, CODASYL-DML transactions on network databases, SQL transactions on relational databases, and Daplex transactions on functional databases. The authors describe MLDS, focusing on its motivation and structure. It is shown how MLDS, by providing an integrated environment for experimenting with data models and data languages, also serves as a testbed that provides insight to data models and data-model semantics, using qualitative and quantitative techniques. Related work on data-language comparison and analysis is indicated. >

Structure memory designs for a database computer

Recent developments in storage technologies have given computer system designers a wide range to consider in building large on-line systems. It is important that the requirements of each component of a proposed system be carefully recognized before choosing a particular technology for its implementation. The design of a database computer which takes advantage of emerging technologies has been proposed. In this paper, the design of an important component of the database computer, namely, the structure memory, is described. Three design alternatives using three different technologies are proposed. These are magnetic bubble memories, charge-coupled devices, and electron beam addressable memories. The relationship of the structure memory with other components of the database computer is briefly outlined.

Performance Evaluation of a Database System in Multiple Backend Configurations

The aim of this performance evaluation is twofold: (1) to devise benchmarking strategies for and apply benchmarking methodologies to the measurement of a prototyped database system in multiple backend configurations, and (2) to verify the performance claims as projected or predicted by the designer and implementor of the mufti-backend database system known as MBDS.

Federated databases and systems: part II --- a tutorial on their resource consolidation

The issues and solutions for the interoperability of a class of heterogeneous databases and their database, systems are expounded in two parts. Part I presented the data-sharing issues in federated databases and systems (Hsiao, 1992). The present article explores resource-consolidation issues.Interoperability in this context refers to data sharing among heterogeneous databases, and to resource consolidation of computer hardware, system software, and support personnel.Resource consolidation requires the presence of a database system architecture which supports the heterogeneous system software, thereby eliminating the need for various computer hardware and support personnel. The class of heterogeneous databases and database systems expounded herein is termedfederated, meaning that they are joined in order to meet certain organizational requirements and because they require their respective application specificities, integrity constraints, and security requirements to be upheld. Federated databases and systems are new. While there are no technological solutions, there has been considerable research towards their development. This tutorial is aimed at exposing the need for such solutions. A taxonomy is introduced in our, review of existing research undertakings and exploratory developments. With this taxonomy, we contrast and compare various approaches to federating databases and systems.

A multi-backend database system for performance gains, capacity growth and hardware upgrade

Traditional database systems have long been plagued by performance problems when there is either an increase in the mainframe usage or in the database applications. Solutions to these problems have been sought, first, by offloading the database system from the mainframe computer to a single, dedicated backend computer. The backend computer has its own disk storage, is used to perform all of the database operations, and interacts with the mainframe. However. database systems with this software single-backend approach still encounter the performance problems when either the backend usage or database applications increase. The software multiple-backend approach to database management and hardware upgrade is therefore proposed to overcome the performance-gains and capacity-growth problems of either traditional mainframe-based database systems or conventional software single-backend database systems. In this paper we specify the design requirements and issues of the software multi-backend database systems. We show how these requirements and issues affect the design and implementation of a multi-backend database system known as MBDS. Since MBDS is designed specifically for performance gains, capacity growth, and hardware upgrade, we benchmark MBDS in order to verify whether its design and implementation can indeed relate the gains and growth directly to the multiplicity of backends in terms of the response-time reduction and invariance.

The multi-lingual database system

This paper introduces a new and unconventional approach to the design and implementation of a database system, the multi-lingual database system (MLDS). The multi-lingual database system is a single database system that can execute many transactions written respectively in different data languages and support many databases structured correspondingly in various data models, i.e., DL/I transactions on hierarchical databases, CODASYL-DML transactions on network databases, SQL transactions on relational databases and Duplex transactions on functional databases. Thus, a multi-lingual database system allows the old transactions and existing databases to be migrated to the new environment, the experienced user to continue to utilize certain favorite features of existing data languages and data models, the new user to explore the strong features of the various data languages and data models, the hardware upgrade to be focused on a single system instead of a heterogeneous collection of database systems, and the database application to cover wider types of transactions and different modes of interactions.

THE FEDERATED DATABASES AND SYSTEM: A NEW GENERATION OF ADVANCED DATABASE SYSTEMS

The presence of large numbers of heterogeneous databases in a given organization is prompted by the replacement of traditional data processing with modern database management systems (DBMS) for traditional applications, such as record keeping, product assemblies and inventory control, as well as for new applications, such as expert-system support, design automation and manufacturing engineering. These heterogeneous databases form a federation in the organization due to the desire to facilitate data sharing and resource consolidation while maintaining the local autonomy of individual databases, which are dictated by their particular security requirements, integrity constraints and application specificities. This paper outlines an architecture for the federated databases, based on the multimodel, multilingual, and multibackend approach, that facilitates data sharing and resource consolidation in a heterogeneous database environment without compromising the autonomy of individual databases.

A methodology for supporting existing CODASYL databases with new database machines

In this paper, an attempt is made to show that conventional database management system software, in particular those of CODASYL type, can be effectively replaced by database machines with good performance. The replacement of CODASYL system software involves two main steps: (i) In order to preserve the notions of CODASYL records, sets,-areas, and others, we need a methodology for database transformation so that an existing CODASYL database may be transformed into suitable formats for storing and retrieving in the database machine. (2) For the purpose of allowing existing application programs written in a CODASYL data sublanguage to store, retrieve and manipulate CODASYL data in the new environment without reprogramming, we need to be able to translate the data sublanguage calls dynamically into the commands of the database machine. Such process is termed query translation.n In this paper, a database transformation methodology and a query translation process are presented which ensure that the content-addressability and parallel read-out capability of the database machine are used to advantage. The machine in consideration is known as the database computer (DBC) and is also briefly reviewed. DBC is one of the typical new back-end machines for database management which utilize the emerging hardware and the modification of existing hardware for performance gain and capacity increase.