A debugging package for use in real-time with the programming language Modula is presented. The authors argue that such a package has some generality and portability in the context of modern high-level realtime langua...
详细信息
A debugging package for use in real-time with the programming language Modula is presented. The authors argue that such a package has some generality and portability in the context of modern high-level realtime languages. The facilities provided by the package, which are at a low level, and their implementation are described. These facilities include a journal, event counting, and event marking. The package has been in use for over one year and some conclusions from this use are presented. However, problems have arisen from the low level and flexible nature of the facilities and these are discussed. Finally, some execution times for the facilities are listed and ideas for future work given.
MINOS is an acronym of MINe Operating System. The rationale which led to the choice of MINOS as a standard computer software system within the UK National Coal Board is discussed. The current status of software engine...
详细信息
MINOS is an acronym of MINe Operating System. The rationale which led to the choice of MINOS as a standard computer software system within the UK National Coal Board is discussed. The current status of software engineering is related to the evolution of an engineering discipline, and the contribution of the reuseable software structure of MINOS to this evolution is studied. An overview of MINOS design considerations is given and applications at UK mines are listed. The reliability and operational safety of MINOS systems is excellent, and it is the design for reuseability which leads not only to reliability but to user confidence in newly installed systems.
This paper briefly considers the kinds of problems encountered in the production of large, real-time systems. These are seen as being of two types : those intrinsic to the system and those connected with the methods a...
详细信息
This paper briefly considers the kinds of problems encountered in the production of large, real-time systems. These are seen as being of two types : those intrinsic to the system and those connected with the methods and tools used during production. Approaches adopted within the Bracknell Division of Ferranti computer Systems Ltd are discussed, with emphasis on a set of testing tools that have been developed. Finally, the impact of Ada 1 is considered briefly.
Errors within an assembler program for limiting the local power density in a reactor vessel are to be identified and corrected during two test phases. According to the special features for error detection of the testi...
详细信息
Errors within an assembler program for limiting the local power density in a reactor vessel are to be identified and corrected during two test phases. According to the special features for error detection of the testing facilities, the errors are classified, and an appropriate software reliability model is applied to each error class. These models allow prediction of certain reliability parameters from the observation of the test. Depending on the model employed these parameters can be e.g. the number of remaining errors, the probability for the occurance of an error within a given time interval, the meantime between successive errors. In this contribution references are only made to the number of remaining errors.
The objective of this paper is to study the tools of on-line development for the decentralized control. In the first part we shall analyse the needs of a decentralized control : architecture of systems based on the ge...
详细信息
The objective of this paper is to study the tools of on-line development for the decentralized control. In the first part we shall analyse the needs of a decentralized control : architecture of systems based on the geographical and hierarchical decomposition ; the stages of the “life” of an application. The second part of this paper will treat the important principles of the on-line perfection and some elementary operations : observation of modules, “prudent trial”, on-line data injection,.... The third part will develop a working example emanating from a particular high-level language of structuration (STR) permitting a simple and efficient realization of these different basic functions.
This paper describes a tool for real time software which can present debugging information with no particular debugging runs or a low degree of repetition of debugging runs. This tool is useful for locating problems i...
详细信息
This paper describes a tool for real time software which can present debugging information with no particular debugging runs or a low degree of repetition of debugging runs. This tool is useful for locating problems i...
详细信息
This paper describes a tool for real time software which can present debugging information with no particular debugging runs or a low degree of repetition of debugging runs. This tool is useful for locating problems in later test phases or after the delivery of large scale real time software which cannot be developed by a single individual and which performs dead line control. Many interface, timing, and performance errors often remain uncovered until later in the life cycles of large scale real time software. In order to determine these problems with conventional tools, it is necessary to repeat debugging runs almost in the same conditions as those when errors occur. Since such repetition of debugging runs is very expensive and because errors are sometimes covered again, it reduces productivity and reliability markedly. The tool proposed here avoids such shortcomings of conventional debugging tools. The tool concept is derived from experience with error analysis obtained during the development of train traffic control software, which is typical of large scale real time software. This tool is expected to considerably improve the productivity and reliability of large scale real time software.
In reports of energy management systems the extent of customer involvement required and the problems associated with the compilation of a reasonably error-free data base and one-line displays are aspects which have re...
详细信息
In reports of energy management systems the extent of customer involvement required and the problems associated with the compilation of a reasonably error-free data base and one-line displays are aspects which have received far too little attention. The effort involved in the building up of an energy management data base is firstly discussed and then quantified. The author makes suggestions on how the magnitude of this task can be reduced. This is followed by a discussion on the problems experienced so far with the operator load flow and associated programs. The basic causes of these problems are identified and some remedies are proposed on how such problems can be eliminated in future projects. Lastly the manpower required to be committed by the customer is discussed.
Process control computers are used in nuclear power plants for very different tasks. Depending on the significance for safety the demands on faultless construction and reliability are more or less urgent. Those demand...
详细信息
Process control computers are used in nuclear power plants for very different tasks. Depending on the significance for safety the demands on faultless construction and reliability are more or less urgent. Those demands are the main topic of the paper. While analysing process control computer systems in nuclear power plants you will find, with respect to the usual separation between hardware and software, some more problems in the field of software than in the hardware. The reason will be discussed. Developments in computer technology may change this situation a little. There are strong efforts to simplify software production, and as a result it may become more easy to check software. Due to increasing integration and hectic changes it will be more and more difficult to obtain full range information about hardware, to process these informations and to verify them with reasonable expense and time.
Data flow dependencies of a program are important for code-optimisation of compilers and for detecting errors in programs. The control flow of a program can be described by its control flow graph. The mutual dependenc...
详细信息
Data flow dependencies of a program are important for code-optimisation of compilers and for detecting errors in programs. The control flow of a program can be described by its control flow graph. The mutual dependencies of variables in a program can be represented by graphs, too. For each variable of a program a tree structure can be found that shows its dependency on other variables. Different kinds of tree structures depict the data flow dependencies more or less detailed, according to the more or less exhaustive use of the source code’s information. The most accurate representation of data flow by tree structures can be achieved for individual paths of a program only. A less accurate representation results from the set of unordered program statements. In this case it is not possible to distinguish whether a statement precedes another or not. By use of the static control flow graph, trees can be established without regard to branching conditions. With these trees distinctions can be made between different actual values of the same variable. The methods of building tree structures that show the data flow of programs are suitable for data flow analysis by hand as well as for automatic analysis.
暂无评论