Get straight to the point of database processing.Database Processing reflects a new teaching method that gets readers straight to the point with its thorough and modern presentation of database processing fundamentals.The twelfth edition has been thoroughly updated to reflect the latest software.
In the preceding chapter we mentioned in passing that many realtime systems, especially those of a process control nature, operated on dedicated computers. Part of the data produced in such dedicated systems may, however, be important input to one or more conventional data processing systems. Such a situation leads naturally to demands to link the dedicated system to the main data processing computer, perhaps initially by means of off-line data communications, but before long by means of on-line or even real time communication.
Almost any organisation has a certain amount of administrative work that must be carried out: staff must be paid, raw materials or finished goods have to be ordered and customers have to be invoiced for items they have purchased. In addition, records have to be maintained of the personnel employed, the tax deducted from their pay, suppliers, orders placed with suppliers but not yet fulfilled, customers, customers' orders that have not yet been satisfied and a hundred and one other items in order that the organisation may work efficiently and meet its legal obligations.
In the previous chapter we have seen that input and output devices are essential parts of any computer used for data processing and also noted that most computers make use of backing store to enhance their capability. Collectively input and output devices connected to the computer together with those types of backing store which are connected to but not fully integrated into the CPU are referred to as peripheral units or more simply peripherals.
Much confusion exists between the terms ‘on-line’ and ‘real time’, which are often used indiscriminately. The two concepts are, however, quite distinct and though a system may be on-line it need not be real time. Conversely, a real-time system must, of necessity, be on-line so the term sometimes seen, ‘on-line real time’ (OLRT), is in fact tautological.
It is one of the paradoxes of automation that it is a labour-intensive activity. That is to say the basic process of identifying, designing and creating automated systems requires considerable effort by people with many different skills. Commercial data processing is no exception to this general statement and in fact adds another dimension to it inasmuch as the operation of a commercial data processing centre also requires a considerable range of skills.
Data communications systems bring together the technologies of data processing and communications engineering. As such these systems require representatives from both disciplines to construct them and make them work. In this chapter a brief overview of the concepts of data communications of most interest to the data processing practitioner is given, and for further details the reader is referred to the appropriate works listed in the Bibliography.
In Chapter 9 the design specification for a new system was described. After approval to proceed has been granted (which should include specific approval by the users of the proposed system as well as technical approval by management services management and resource allocation approval by senior management) it is the analyst's responsibility to produce the detailed design for a workable system.
The computers used in data processing are composed of one or more electronic machines using digital signals for recording and moving data and instructions.
No matter how good the design work and programming that has been carried out, or how good a functional fit a purchased package may be, no system makes any contribution towards the effectiveness of an organisation until it has been brought into use. This process is a difficult one since systems are rarely installed in 'green field sites' and careful transition from an existing system or set of procedures to the new system has to be achieved.
The initial applications of computers to commercial data processing were all of a batch nature and although the other processing modes have gained in importance much of the routine work in most installations is still performed in this way. Nor can this be considered as merely a legacy from a less-sophisticated age of data processing since for some types of application batch processing is the natural and most effective approach.
