Data-Processing System

a system of interrelated techniques and means of collecting and processing the data needed to organize control of some system. Automatic, or electronic, data-processing systems make use of electronic computers or other modern information-processing equipment. Without a computer a data-processing system can be constructed for only a small controlled system. The use of a computer means that the data-processing system can perform not just individual information-processing and computing tasks but a set of tasks that are interconnected and can be carried out in a single sequence of operations.

Data-processing systems should be distinguished from automated control systems. The primary function of the latter is the performance of calculations associated, for example, with the solution of problems of control and with the selection of optimal variants of plans on the basis of models and the techniques of mathematical economics. The chief purpose of automated control systems is to increase efficiency of control or management. The functions of data-processing systems, however, are to collect, store, retrieve, and process the data needed to carry out the calculations at the lowest possible cost. When an automatic data-processing system is constructed, efforts are made to identify and automate laborious, regularly repeating routine operations on large files of data. A data-processing system is usually a part of an automated control system and represents the first stage in the development of an automated control system. Data-processing systems, however, also function as independent systems. In many cases it is more efficient to use a single system to process similar data for a large number of control problems handled by different automated control systems —that is, to use a shared-access data-processing system.

The first data-processing systems were constructed in the USA in the 1950’s, when it became clear that the use of a computer to solve individual problems, such as calculating wages or keeping track of stocks of goods and materials, was inefficient. It was seen at that time that integrated processing of the data fed to the computer was necessary.

The USSR has a number of large data-processing systems, most of which are the bases of automated control systems. Examples are the systems that have been set up at such large industrial enterprises as Frezer, Kalibr, the Likhachev Automotive Plant, the L’vov Television Plant, and the 15th Anniversary of the Ukrainian Lenin Komsomol Donetsk Plant. Data-processing systems are coming into use not only in industrial enterprises but also in planning bodies, statistical agencies, ministries, and banking institutions. They are finding application in trade and in the supply of materials and equipment. The introduction of a data-processing system is a prerequisite for the development of an automated control system.

The experience that has been gained with data-processing systems permits identification of the basic principles of the construction and techniques of development of such systems. The most important principle is the principle of integration, which requires that the raw data undergoing processing be fed to the data-processing system once. The problems being solved in the data-processing system are coordinated in such a way that the raw data and the data resulting from the solution of some problems are used as the initial data for as many of the other problems as possible. This coordination eliminates duplication of the operations of collection, preparation, and checking of data and ensures the integrated use of the data. As a result, the costs of obtaining the necessary information are reduced, and the efficiency of the data-processing system is increased.

Closely related to the principle of integration is the principle of centralization of data processing. When a data-processing system is constructed, many information-processing tasks are removed from the control of the respective subdivisions and are concentrated at a single computing center or at a small number of such centers. Large data files are established at these centers; the files are available for integrated processing. Special information retrieval systems, called automatic data banks, are set up in the data-processing system to manage and make optimal use of the files. The automatic data bank receives data that are subject to repeated use, and, in conformity with the operating schedule of the data-processing system, the data are used to form the work files for the problems being solved. The data bank also supplies information in response to inquiries. The centralization of data processing in constructing a data-processing system usually assumes a reorganization of the structure of control.

The principle of the systems approach to the organization of the sequence of operations consists in the following: When the data-processing system is constructed, there must be integrated mechanization and automation of the operations at all stages of data collection and processing, and the hardware used must be self-consistent with respect to throughput and other parameters. If this is not done, the unity of the sequence of operations is disrupted, and the efficiency of the data-processing system drops sharply.

Before a data-processing system is constructed, the following are subjected to thorough investigation and analysis: the controlled system, the control problems, the structure of control, the content of the information, and the information flows. On the basis of the analysis of the results of the investigation, an information model of the data-processing system is developed that establishes new information flows and the relation between the data-processing tasks. The hardware is chosen, and the sequence of operations of the data-processing system is worked out, according to the volumes of data being processed, stored, and transmitted as determined from the information model of the data-processing system. The successful construction of a data-processing system requires the participation not only of specialists but also of managers and other personnel who are directly involved in the solution of control problems at all stages of the development and introduction of the system.