Self-metric software are programs that record their behavior statistics automatically during their execution. They are generated from existing software by inserting software measurement instruments into the program at...
详细信息
Self-metric software are programs that record their behavior statistics automatically during their execution. They are generated from existing software by inserting software measurement instruments into the program at appropriate locations after an automatic analysis by the system. An algorithm to locate and insert the optimal (minimal cost) set of activity counters for generating the program activity profile is given. The use of the model for measuring execution time, branching characteristics, code activities, path activities, and other program statistics are discussed. The applications of these statistics in program testing, optimization, and program restructuring for virtual memory are presented.
Report on procedure in which the human factors input is made at the stage at which system planning is under way, when system characteristics are sufficiently well defined that it is possible to write a dialog specific...
详细信息
Report on procedure in which the human factors input is made at the stage at which system planning is under way, when system characteristics are sufficiently well defined that it is possible to write a dialog specification, but before systems designers have begun to write dialogs. In conjunction with the development of a large-scale man-computer system proposed by the Agricultural Stabilization and Conservation Service such a set of dialog specifications is written. The specifications permit several system design teams to work independently, but produce dialog that is relatively uniform from subsytem to subsystem and that meets standards for good human factors design. The steps undertaken to create the specification are described together with selected samples of the kind of dialog that may result.
A career development program for information systems practitioners currently being used widely by employers in the UK and now becoming available in North America is described. The program, called the Professional Deve...
详细信息
A career development program for information systems practitioners currently being used widely by employers in the UK and now becoming available in North America is described. The program, called the Professional Development Scheme (PDS), was developed by the British computer Society to address the lack of structure and quality control generally present in the way computing professionals were being trained. The performance standards underpinning the program (The British computer Society Industry Structure Model) have recently been thoroughly updated and now include material specific to the development, maintenance, and management of software for safety-critical applications. The use of the program for this purpose and potential developments in the field of training and registration for safety-critical software specialists are discussed.
A queueing analysis is presented of a pre-emptive, priority driven, timesliced dispatcher algorithm, typical of those found in many timesharing systems. A distinctive feature of the system being modeled is that a pree...
详细信息
A queueing analysis is presented of a pre-emptive, priority driven, timesliced dispatcher algorithm, typical of those found in many timesharing systems. A distinctive feature of the system being modeled is that a preempted task, when re-admitted to the dispatcher, is re-allocated a full timeslice rather than the residual. The combination of pre-emption, timeslice renewal and class dependent arrival and service rates, places the analysis beyond that of the well-known FB//N algorithms. The model is used to predict the behavior of the dispatcher algorithm under various workloads.
The increasing complexity of computer and communication systems in use today has raised the levels of risks of exposure to failure and accidents. One approach to address this problem is by reducing the incidence of un...
详细信息
The increasing complexity of computer and communication systems in use today has raised the levels of risks of exposure to failure and accidents. One approach to address this problem is by reducing the incidence of unnecessary complexities which stem from the mapping of application functions onto poor or inappropriate execution platforms. The unnecessary complexities may be addressed by restructuring the hardware and software architecture complex. In addition to this, disciplined structuring and execution of life cycle processes with the support of appropriate methods and tools are needed.
An in-place reversible sorting algorithm is used following a fast Walsh transform to sort coefficients into increasing sals followed by decreasing cals. This separation of the coefficients into those arising from odd ...
详细信息
An in-place reversible sorting algorithm is used following a fast Walsh transform to sort coefficients into increasing sals followed by decreasing cals. This separation of the coefficients into those arising from odd sinelike time functions and those from even cosinelike functions is useful in signal processing applications. A second sort routine generates a relocating set for a remap sequency sorting. To be valid for all N this sort routine is done out-of-place. The separation of the transform and sort functions has resulted in simple implementations for each. To demonstrate the differences in the sort routines, transform results and their inverses are made on a linear ramp. For this application the transform results with sal-cal sorting has all zeros beyond the mid-point showing that, with the exception of the large d. c. term, the linear ramp is composed entirely of Walsh odd time functions.
The need for formal proofs is much more crucial for parallel processes than for sequential ones. In this cconnection, S. Owicki has developed an interesting method for proving partial correctness of parallel programs....
详细信息
The need for formal proofs is much more crucial for parallel processes than for sequential ones. In this cconnection, S. Owicki has developed an interesting method for proving partial correctness of parallel programs. This same technique may also be used to exhibit the behavior characteristics of non-terminating processes. A check is made of this technique's efficiency by applying it to a problem already used to compare various synchronization techniques. This permits an analysis to be made of the power of the method. Comments are made on its applicability and some suggestions for future work are presented.
A methodology using attributed grammars representing the formal specification of distributed computing system software design is developed. This methodology provides a formal expression for synchronization, concurrent...
详细信息
A methodology using attributed grammars representing the formal specification of distributed computing system software design is developed. This methodology provides a formal expression for synchronization, concurrent processing and interprocess communication in the design of distributed computing system software using productions of attributed grammars. These productions can express sequencing, selection, recursion and concurrency in the system's activities, and can be used for expressing stepwise refinement of the design. The control flow of the software system is governed by the productions and the attributes of the symbols in the productions in the grammar. The formal specification of distributed computing system software design can also be used for consistency checking of the design and validation of implementation at run-time. An example is given to demonstrate this methodology.
Program errors detected during internal testing of the operating system DOS/VS form the basis for an investigation of error distributions in system programs. Using a classification of the errors according to various a...
详细信息
Program errors detected during internal testing of the operating system DOS/VS form the basis for an investigation of error distributions in system programs. Using a classification of the errors according to various attributes, conclusions can be drawn concerning the possible causes of these errors. The information thus obtained is applied in a discussion of the most effective methods for the detection and prevention of errors.
The objective of this study is to quantify in terms of the results of an analytic model the effect wait list positioning can have on system performance. In systems where CPU or channel time is a bottleneck, the model ...
详细信息
The objective of this study is to quantify in terms of the results of an analytic model the effect wait list positioning can have on system performance. In systems where CPU or channel time is a bottleneck, the model can be used to study the importance of wait list positioning in increasing or decreasing job class response. The analytic model of different wait list positioning strategies is first presented. Performance predictions of the positioning policies are discussed and related to system and job class performance.
暂无评论