The proceedings contains 205 papers. It consists of 3 volumes entitled: architecture, decision support systems, and knowledge-based systems;software;and healthcare systems, and special topics. The topics discussed inc...
详细信息
The proceedings contains 205 papers. It consists of 3 volumes entitled: architecture, decision support systems, and knowledge-based systems;software;and healthcare systems, and special topics. The topics discussed include: system implementation in VLSI;GaAs processor design and related issues;systolic/pipelined parallel architectures;knowledge acquisition for expert systems;group decision support systems;code analysis and maintainability;compiler technology;distributed database systems;medical image analysis/radiology systems;medical decision support systems;hospital systems;biomedical applications;and special topics (such as automated legal analysis, financial accounting databases for MIS, and safeguarding information systems technology).
The purpose of this work is to analyze the applicability of artificial intelligence techniques for developing a user-friendly, parallel interface to large isolated, incompatible NASA databases for the purpose of assis...
详细信息
The proceedings contain 29 papers. The topics discussed include: storage management in a prolog compiler;developing application independent interfaces for workstations in a distributed environment;moving decision poin...
ISBN:
(纸本)089791211X
The proceedings contain 29 papers. The topics discussed include: storage management in a prolog compiler;developing application independent interfaces for workstations in a distributed environment;moving decision points outward from applications and utilities and into command level;performance characteristics of microcomputer prolog implementations;a C language extension for machine-independent programming;KOALA : a cost effective workstation for fast lisp interpretation;recent developments in PC software;viewing messages as objects;reliable commit and optimistic concurrency control for dynamically reconfigurable distributeddatabases;remote operations across a network of small computers;parallel computing on personal computers;a stream-oriented approach to distributed query processing in a local area network;a packet voice communications over pc based local area networks;a practical, distributed environment for Macintosh software development;languages for expert system building: a comparison;and SFM: synchronizing forms manager.
Dragon Slayer is a distributed operating system for microprocessor-based networks. It is currently implemented on a carrier-sense multiple-access (CSMA) network of 16 leading edge PCs. It provides message passing and ...
详细信息
ISBN:
(纸本)0818607246
Dragon Slayer is a distributed operating system for microprocessor-based networks. It is currently implemented on a carrier-sense multiple-access (CSMA) network of 16 leading edge PCs. It provides message passing and uses a standard client/server model. A unique aspect of the system is the lack of any global process existence, global system kernel or central authority. The authors explain their use of broadcast messaging and the mechanism for consistent and unique process naming. They achieve fair resource scheduling without any assumption on global management or control using a novel distributed algorithm derived from constructions using the theory of interaction systems. Further, they address the usefulness of Dragon Slayer for dealing with data integrity in distributeddatabases, for distributed vision processing applications, and for providing cheap, efficient, robust solutions for real-time process control problems in automatic manufacturing.
Concern for performance of concurrency control protocols, in distributed database systems, raises the question whether performance is improved by permitting intratransaction concurrency. This question is answered for ...
详细信息
ISBN:
(纸本)0818607246
Concern for performance of concurrency control protocols, in distributed database systems, raises the question whether performance is improved by permitting intratransaction concurrency. This question is answered for protocols which use abortion of transactions to preserve consistency. Variants of the protocol proposed by D. P. Reed (1978) are simulated and used to represent protocols which abort transactions. The results indicate that as the workload increases the intratransaction concurrency leads to poorer performance. Further, they suggest that all entities should be assigned an order and, whenever the access set of a transaction is known a priori, the transaction should access entities in this assigned order with possible skips.
The following topics are dealt with: global fault-tolerant system design;design for testability;parallelsystems;field experience of computer controlled switching systems;system aspects of fault-tolerant distributed s...
详细信息
ISBN:
(纸本)0818607033
The following topics are dealt with: global fault-tolerant system design;design for testability;parallelsystems;field experience of computer controlled switching systems;system aspects of fault-tolerant distributedsystems;testing;fault-tolerant hardware architectures;methods and tools for evaluation;interconnection/communication networks;backward error recovery;method of considering time-dependent-faults;experimental evaluation;artificial intelligence and fault tolerance;experimental evaluation of software errors and failures;fault masking to multiple systems;built-in self tests;safety issues for computer controlled systems;reliable broadcasting and clock synchronization;theory of system level diagnosis;and dependability modeling and measurements. 63 papers and a panel session summary are published in the present proceedings.
This conference proceedings contains 466 papers by various authors. The following topics are dealt with: layout for VLSI;CAD for super computers;digital filters;active circuits;parallel and distributed computation;bro...
详细信息
This conference proceedings contains 466 papers by various authors. The following topics are dealt with: layout for VLSI;CAD for super computers;digital filters;active circuits;parallel and distributed computation;broadband matching;nonlinear problems and chaos;large-scale simulation;VLSI filtering techniques;approximation for filters;active filters;coding system;system analysis and estimation;nonlinear circuit analysis;VLSI synthesis methods;modeling and simulation;design and methodologies for VLSI implementation;discrete-event systems;switched-capacitor networks;analysis and diagnosis of analog circuits;oscillators;nonlinear circuits and systems;algorithms for gate-level ATPG;information network;digital signal processors;graph theory with applications;chaos, bifurcations, and related topics;VLSI design;CAD and computer graphics in Japan;modeling and simulation of submicron devices, circuits, and systems;combinatorial algorithms and optimization;power electronics;layout algorithms for VLSI;VLSI circuits simulation;2-D signal processing;adaptive signal processing;graph theory;analog filters;power systems;VLSI testing;image processing;multivariable networks and systems;continuous-time integrated filters;communication circuits;logic circuits;multidimensional signal processing;microwave filters and multiplexing networks;adaptive filters and systems;and parallel processing.
proceedings incorporates 32 papers that are subdivided into eight groups dealing with the following subjects: protocol implementation, micro systems, integrated office systems, network monitoring and control, distribu...
详细信息
ISBN:
(纸本)0444877150
proceedings incorporates 32 papers that are subdivided into eight groups dealing with the following subjects: protocol implementation, micro systems, integrated office systems, network monitoring and control, distributeddatabases, local area networks (LAN's), performance evaluation, and network architecture. Although most of the presented articles are of theoretical nature, some results of practical value are reported. Topics considered include: packet-switching networks, OSI protocols, color graphics, office microcomputers, man-machine dialogue, word and text processing, distributed programming, ALOHA channels, database systems, computer architecture, control algorithms, interoffice mail system, and data communication.
An approach is proposed for modeling off-the-shelf hardware and for modeling parallel algorithms, along with a design methodology to use the information provided by these models, to design a class of macro-pipelined s...
详细信息
ISBN:
(纸本)0818606347
An approach is proposed for modeling off-the-shelf hardware and for modeling parallel algorithms, along with a design methodology to use the information provided by these models, to design a class of macro-pipelined special-purpose architectures. Nine parameters to form a model of the characteristics of parallel/distributed algorithms and the environment in which they must execute are presented. In addition, a set of tuples to model the characteristics of computer architectures is presented. By combining the tuples with the parameters, the execution time of the algorithm modeled by the parameters on the hardware modeled by the tuples can be approximated. The combination of these models could be used as a basis for computer-aided tools used in the design of macro-pipelined parallel/distributed processors.
MU6V is a parallel vector processing system in which a linear set of processors, each capable of both vector and scalar operations, is interconnected by a common communication highway. Memory is fully distributed amon...
详细信息
ISBN:
(纸本)0818606347
MU6V is a parallel vector processing system in which a linear set of processors, each capable of both vector and scalar operations, is interconnected by a common communication highway. Memory is fully distributed among the processors and interprocessor communication occurs by means of global broadcasts controlled in hardware by a data-driven synchronization mechanism. A prototype model based on 68000 microprocessors, in which vector orders are emulated as 'extracodes,' has been constructed. A host machine provides a software support environment and a run-time system in the prototype provides diagnostic information. Programming language features have been designed to exploit this configuration and an example algorithm is presented.
暂无评论