The economic production of reliable computer system software requires an integrated set of management and development procedures, standards, and tools. A responsive methodology and software tools set is described, alo...
详细信息
The economic production of reliable computer system software requires an integrated set of management and development procedures, standards, and tools. A responsive methodology and software tools set is described, along with several unresolved software development issues and areas for methodological improvement.
Despite the widely acknowledged importance of reader feedback in the assessment of application software documentation, few studies have sought to establish what subjective factors readers use when evaluating documenta...
详细信息
Despite the widely acknowledged importance of reader feedback in the assessment of application software documentation, few studies have sought to establish what subjective factors readers use when evaluating documentation and how to reliably measure those factors. This paper reports the development and validation of an instrument which measures seven important reader-oriented factors: credibility, demonstrative, fitness, personal affect, systematic arrangement, task relevance, and understandability.
In modern I/O architectures, multiple disk drives are attached to each I/O controller. A study of the performance of such architectures under I/O-intensive workloads has revealed a performance impairment that results ...
详细信息
In modern I/O architectures, multiple disk drives are attached to each I/O controller. A study of the performance of such architectures under I/O-intensive workloads has revealed a performance impairment that results from a previously unknown form of convoy behavior in disk I/O. In this paper, we describe measurements of the read performance of multiple disks that share a SCSI bus under a heavy workload, and develop and validate formulas that accurately characterize the observed performance (to within 12% on several platforms for I/O sizes in the range 16-128 KB). Two terms in the formula clearly characterize the lost performance seen in our experiments. We describe techniques to deal with the performance impairment, via user-level work-arounds that achieve greater overlap of bus transfers with disk seeks, and that increase the percentage of transfers that occur at the full bus bandwidth rather than at the lower bandwidth of a disk head. Experiments show bandwidth improvements of 10-20% when using these user-level techniques, but only in the case of large I/Os.
By restricting attention to a particular class of parallel algorithms associated with the divide-and-conquer principle, and choosing an appropriate network topology, the authors have developed an attractive scheme whi...
详细信息
By restricting attention to a particular class of parallel algorithms associated with the divide-and-conquer principle, and choosing an appropriate network topology, the authors have developed an attractive scheme which they call ZAPP (Zero Assignment Parallel Processor) for 'buying speed' from VLSI in a programmable manner. The scheme supports variable grain exploitation of parallelism via a 'lazyflow' execution model, and appears particularly well suited for developing a high-performance personal computer which realizes the inherent parallelism in applicative languages.
Algebraic specification technique for the description of data types is considered. It consists of two parts: (1) the syntactic specification providing a finite set of operation symbols together with domain and range i...
详细信息
Algebraic specification technique for the description of data types is considered. It consists of two parts: (1) the syntactic specification providing a finite set of operation symbols together with domain and range information; (2) the semantic specification consisting of a finite set of ″equations″ built from operation symbols, constants, and variables. The technique bears a number of problems which are discussed in the paper.
The GD80 graphic display family with its modular architecture has been designed to cover a large range of CAD applications. Five microprocessors are dedicated for different tasks, peripheral handling, picture processi...
详细信息
The GD80 graphic display family with its modular architecture has been designed to cover a large range of CAD applications. Five microprocessors are dedicated for different tasks, peripheral handling, picture processing and generation, application programming and communication with a host. Different multiprocessing methods: common bus for data transfer and interprocessor interrupt, memory sharing with mutual exclusion, dual-port memories, FIFO-s, bus windows and microprogrammed multiprocessing are used for interprocessor communication. Standard 8 bit microprocessors and high speed, horizontally microprogrammed special processors are the basic building blocks of the graphic system described.
0(k log N) algorithms are obtained for permuting and sorting N data items on cube and perfect shuffle computers with N**1** plus **1**/**k processing elements, 1 less than equivalent to k less than equivalent to log N...
详细信息
0(k log N) algorithms are obtained for permuting and sorting N data items on cube and perfect shuffle computers with N**1** plus **1**/**k processing elements, 1 less than equivalent to k less than equivalent to log N. These algorithms directly lead to a generalized-connection-network construction having 0(k log N) delay and 0(k N**1** plus **1**/**k log N) contact pairs. This network has the advantage that the switches can be set in 0(k log N) time by either a cube or perfect shuffle computer with N**1** plus **1**/**k processing elements.
Teaching computer science and particularly programming in the curriculum of a mechanical engineering bachelor poses a difficult task due to motivational factors of the students. In this paper we describe the measures ...
详细信息
To date the necessary rationalization procedures in the manufacturing environment to achieve competitive advantage have taken place mostly from a high level, centralized point of view within organizations. Additional ...
详细信息
To date the necessary rationalization procedures in the manufacturing environment to achieve competitive advantage have taken place mostly from a high level, centralized point of view within organizations. Additional competitive success can only be achieved through consequential shop floor orientation. Existing resources at the shop floor level, for example, processing proximity and manufacturing know-how can be used to an advantage. In this paper, a shop floor oriented approach is presented using the example of reduction of the run-in time of CNC-Programmes. A computer aided method for the run-in of CNC-Programmes is presented. Up to now this task has been undertaken empirically. The result is a transparent, well structured and thus target oriented run-in system which assists the production planning process and which leads to significant time saving.
Evaluation of a computer system is an art which relies upon an analysis of pertinent data. A methodology is described which allows data relating to the behavior of a system to be collected by software without introduc...
详细信息
Evaluation of a computer system is an art which relies upon an analysis of pertinent data. A methodology is described which allows data relating to the behavior of a system to be collected by software without introducing perturbation. Simulation techniques which allow the system to be represented by a real operating system in a simulated environment, permit performance estimates of simulated hardware configurations under the influence of bench mark programs.
暂无评论