摘要:

The paper examines the trade-off between cost and performance of a hypercube based computer interconnection network founded on a detailed cost analysis model. A cost function is proposed which considers the economies of scale of bandwidth available through optical fibre communication and the nonincremental increase in node complexity resulting from variations in network size. When considering topology choice, an important consideration is expansion. Any comparison of expansion should be done through marginal analysis: weighing the additional benefits (performance) against the additional cost resulting from expansion. To demonstrate this technique, four interconnection networks are considered: booleann-cube, nearest neighbour mesh hypercube, generalised hypercube, and spanning multiaccess channel hypercube. Systems with equalised performance are studied, investigating the variations in cost. This permits the investigation of the trade-off of performance/cost between high capacity communication channels and increased number of channels and degree.

DOI:10.1049/ip-e.1989.0002

出版商:IEE    

年代:1989

数据来源: IET

摘要:

A tri-state map is developed as a general format for Karnaugh maps, Reed-Muller coefficient maps and transition (partially polarised) maps. The polarisation statuses of the variables in transition maps and Reed-Muller coefficient maps are indicated by the labels on the tri-state maps. A variable in a tri-state map can exist in two of three different states: true, complemented or nonexistent. Karnaugh maps and positive-polarity or negative-polarity Reed-Muller coefficient maps are special cases, in which there are only two states. Minimisation of exclusive-OR switching functions can be carried out, not only on Reed-Muller coefficient maps, but also on Karnaugh maps or any of the transition maps.

DOI:10.1049/ip-e.1989.0003

出版商:IEE    

年代:1989

数据来源: IET

摘要:

The simulation at component level of microprocessor systems provides a precise technique for evaluating the design of a system with regards to its requirements specification. The paper describes the use of occam to simulate individual microprocessor components, and presents a method of modelling the behaviour of microprocessor bus systems. The possibility of utilising transputers to provide real-time simulation is discussed.

DOI:10.1049/ip-e.1989.0004

出版商:IEE    

年代:1989

数据来源: IET

摘要:

A method is described for the identification of sticking bits on a data transmission bus. The method is probabilistic in nature and can use on-line bus data. It is also suitable for use against intermittent faults. Exact and useful approximate analyses are given for various diagnostic procedures using overall bus data words and compartmentalised bus data words. Numerical performance data are given for each procedure.

DOI:10.1049/ip-e.1989.0005

出版商:IEE    

年代:1989

数据来源: IET

摘要:

This paper considers content addressability and discusses its potential impact on software engineering in the context of its affinity to, and thus efficient implementation of, common abstract data types. More generally, we present an approach to language design based on the abstracted characteristics of the implementing hardware. Based on the perceived needs of common data types, a prototype MOS VLSI content addressable memory component is designed together with enhancements leading to large fault-tolerant memory systems. A novel computer architecture is introduced to include this component and to provide an example target machine. The specification of a high level programming language, using a tractable set of extensions to conventional language syntax, then completes the programming environment.

DOI:10.1049/ip-e.1989.0006

出版商:IEE    

年代:1989

数据来源: IET

摘要:

In the increasingly vital area of digital network testing, there have been proposals for both input compaction and output compaction using linear Ex-OR functions. In the paper we examine these proposals using the spectral domain rather than the conventional Boolean domain, and indicate that further research may be profitable.

DOI:10.1049/ip-e.1989.0007

出版商:IEE    

年代:1989

数据来源: IET

摘要:

This paper considers autocorrelation testing for the detection of single stuck-at faults on non-syndrome testable input lines of an internally unate combinational network. The question of an autocorrelation test covering more than one input fault is considered. Also a test circuitry which requires only 2ninput assignments for ann-input system is presented. The work of this paper is an extension of previous work on autocorrelation testing by Eris and Muzio [6].

DOI:10.1049/ip-e.1989.0008

出版商:IEE    

年代:1989

数据来源: IET

摘要:

The paper examines tradeoffs in the design of real-time electronic hardware for automated inspection. Starting with data from an earlier paper, it develops a methodology for optimising hardware implementations. Cost/time ratios of proposed hardware modules and the cost * time product of the whole system are found to be important and related criterion functions. Overhead costs and times complicate the picture, but a procedure is obtained that solves this problem. The emphasis of the work is on removing randomness in the design of hardware systems.

DOI:10.1049/ip-e.1989.0009

出版商:IEE    

年代:1989

数据来源: IET

摘要:

The direct digital multiplication of complex numbers is examined. The proposed technique bypasses the classical method based on the identity (A+Bj)(C+Dj) =AC−BD+ (AD+BC)j. The complex multiplication is considered as one operation and is examined in the bit-level. Various multipliers, such as the parallel or serial-parallel types and the realisation through microprogramming on an arithmetic unit, are presented, all of which are based on the above technique.

DOI:10.1049/ip-e.1989.0010

出版商:IEE    

年代:1989

数据来源: IET

摘要:

When conventional CSMA/CD methods are applied in multichannel environments, the total system capacity is often dependent upon the static and dynamic channel assignment of user stations. If this is the case, user performance may suffer drastically since the traffic-handling capability of the system fluctuates with the system traffic (generation) statistics. The CSMA/CD/IC protocol proposed by Marsan and Roffinella does not have these shortcomings, because channels are selected for use in a dynamic fashion which follows the instantaneous system load. Unfortunately, CSMA/CD/IC requires that all stations possess a separate receiver for each channel. In the paper, we investigate the possibility of using a single receiver per station and yet achieve the automatic global load-balancing capability of the CSMA/CD/IC protocol. It is found that although both persistent and nonpersistent versions of MICON performed as well as CSMA/CD/IC in our tests, MICON imposes different restrictions on the system. Specifically, MICON exploits the fact that in typical network environments, complete connectivity between all stations at any time instant is not required. As a result, MICON requires additional station synchronisation and permits full connectivity only between dynamically established sets of user stations.

DOI:10.1049/ip-e.1989.0011

出版商:IEE    

年代:1989

数据来源: IET