Jay-Louise Weldon

An Integrated Approach to Database Design

This paper provides an integrated approach for research related to the problem of database design. The process of database design is classified into five phases: requirements analysis, view modeling, view integration, view restructuring, and schema analysis and mapping. The input, processing steps, and output for each phase are described. The problems associated with each phase are pointed out. Existing approaches to database design are reviewed and related to these five phases. The significance of this integrated approach for the development of computer-aided methodologies for database design is discussed.

Implementation strategies for the census data base

The decisions which must be made during the implementation of a very large data base may be classified into three major categories: decisions related to data base organization, those concerning the selection and use of data storage devices, and those related to data compression. The logical organization of the data base must be determined to suit data base users and a corresponding physical organization must be selected to implement that logical organization in an efficient manner. The type and number (size) of storage devices to be used must be determined. The nature of these devices and whether the storage bank will be homogeneous or consist of a mix of device types must also be determined, Finally, the feasibility of data compression must be investigated. If desirable, an efficient method must be selected and the extent of its application determined.

Probability distributions and search schemes

Abstract One property of search schemes used frequently as an indication of search efficiency is the expected number of passes (or read operations) required to locate a record in computer memory (or secondary storage). However, other properties related to the variation in the number of passes required may be equally important in determining the most efficient search technique for a given file (or set of records). These properties can be obtained from the distribution of the number of passes. In this paper, the binary search scheme is discussed within this framework. The probability distribution of the number of passes required is derived along with the mean, standard deviation and the percentiles. The mean and standard deviation of the random sequential search scheme are also presented. The application of these results to common search problems in record access and retrieval are discussed and possible extensions involving a combination of binary and random search techniques are indicated.

Logical Data Base Design

The purpose of a data base is to represent objects and events of interest to the organization. Any representation of the real world is, of necessity, incomplete in that it is not the actual object but a description of that object. The characteristics described are those that are most obvious, or most interesting, or most relevant to the creator of the representation. Aspects which are not of interest are ignored. For example, the descriptive characteristics of an individual that are recorded on a driver’s license include age, hair color, eye color, name, and address but not marital status or educational history.

The Data Base Design Process

The objective of data base design is to produce an integrated data base which is accurate and secure and which supports application systems in an efficient manner. Each of these characteristics—integration, integrity, security, and performance—must be addressed during the design process to assure the desired outcome. Integration is the result of specifying data base content and structure in response to the needs of more than one application area. Integrity is assured through controls on application systems that prevent data loss or improper modification. The use of procedures and dbms features that promote data sharing, while limiting access to data which are not shared, results in data base security. Finally, data base performance depends on both the efficiency of the physical realization of the data and the ability of the data base to satisfy processing requests in a timely manner.

The DBA in Practice

The discussions of dba roles and responsibilities in the previous chapters may give the impression that dba is a formal and well-structured function in all organizations where it exists. In actuality, this is not so. The broad and detailed view presented so far is meant to touch on the many aspects of the dba function. However, one must realize that the full dba function is seldom if ever encountered in practice.

DBA Organization and Staff

The dba group is the organizational subunit charged with carrying out the tasks and responsibilities of data base administration. Typically the dba group reports within the data processing department and draws its staff primarily from that department. Reflecting both its organizational position and internal policies on naming, the dba may be called Manager of Data Base Administration, Director of Data Base Services, Manager of Information Resources, or any of a number of similar titles. Where no dba group exists, the administrative and technical tasks of data base administration are assigned to existing groups within edp. Technical tasks are performed by groups within the computer operations or system services department, sometimes called dbms Support, Data Base Design, On-line Systems. Administrative tasks are assumed largely by data processing management.(1)

The DBA Function

The data base concept offers many advantages to organizations that adopt it. The pooling of formerly diverse data files into an integrated data base can reduce redundancy of data and promote consistency in the reports and processes that use the data base. The use of generalized software to store and access the data base can reduce the duplication in application programming and system development that existed when these tasks were the province of each individual system. Finally, the existence and use of dbms-related software, such as query languages and data dictionaries, makes the data base more available and more accessible to both programmers and end users of the data.

DBA within the Organization

The activities of data base administration require that the DBA interact with several different groups, both within and outside the organization. These interactions and the dba’S role in each affect the proper location for the dba function within the organization. Thus, to understand the impact of alternative positions for dba within an organization, one must first understand the nature of these interfaces.

Data Base Standards

Standards are rules and procedures established by an authority in an attempt to measure the quantity or quality of some object or process. In data processing, standards are promulgated by edp management to measure the quality of data processing systems and their operation. The variety of standards that may be developed for a data processing installation cover the full range of systems activities(1) Standards may exist for system development (including project management and the specification of system components, such as backup and recovery procedures), for data and file definition, for program coding, for all aspects of system documentation (including manual procedures and user training), for system test and conversion, and finally for system operation (including the specification of the hardware/software environment of the installation itself).