By Topic

Distributed Computing Systems in the 1990s, 1988. Proceedings., Workshop on the Future Trends of

Date 14-16 Sept. 1988

Filter Results

Displaying Results 1 - 25 of 64
  • Proceedings: Workshop on the Future Trends of Distributed Computing Systems in the 1990s (Cat. No.88TH0228-7)

    Publication Year: 1988
    Request permission for commercial reuse | PDF file iconPDF (192 KB)
    Freely Available from IEEE
  • Mutual exclusion of locally byzantine processes

    Publication Year: 1988, Page(s):232 - 236
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (332 KB)

    A solution of the mutual exclusion problem for processes that can have byzantine behavior, except for particular sections of their code, is presented. The problem is to design a protocol to ensure that, at a given time, no more than one process is executing in its own critical section. The protocol must yield some fairness, absence of starvation, and deadlock requirements. It is assumed that distr... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A hierarchical message mechanism for distributed systems software

    Publication Year: 1988, Page(s):132 - 138
    Cited by:  Patents (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (680 KB)

    The authors investigate a mechanism for distributed computer systems that can support control and data communication right from the beginning of a distributed-software design to the actual execution and maintenance. The concept of message hierarchy is illustrated by an example of remote rendezvous in distributed systems. The control messages need a faster communication structure than the data mess... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A global load balancing strategy for a distributed computer system

    Publication Year: 1988, Page(s):93 - 102
    Cited by:  Papers (3)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (1052 KB)

    A load-balancing strategy is described for a computer systems connected by a multiaccess broadcast network. The strategy uses the existing broadcast capability of these networks to implement an efficient search technique for finding stations with the maximum and the minimum workload. The overhead of distributing status information in the proposed strategy is independent of the number of stations, ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Studying the use and application behaviour of a large network

    Publication Year: 1988, Page(s):14 - 22
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (608 KB)

    A session-level network monitor that collects network application and user information is described. Such information can potentially be used for various network-management functions, including performance, configuration, accounting, and security management. The design of a prototype monitor for the DECnet environment is reviewed. The use of the monitor is illustrated by applying it to monitor a s... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Migrating transactions

    Publication Year: 1988, Page(s):512 - 520
    Cited by:  Papers (3)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (448 KB)

    Presents a concept that permits long activities to be broken into several steps using short transactions. The scheme maintains the consistency of the data and the correct termination of the transaction-set is guaranteed. Invariant data or program structures can be passed to subsequent decision points, permitting an application designer to specify those assumptions needed to produce the actions tha... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A controlled learning environment of enhanced perceptron

    Publication Year: 1988, Page(s):452 - 460
    Cited by:  Papers (6)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (544 KB)

    Some basic features of multi-layer perceptrons are sketched. Some of the shortcomings of these perceptrons are pointed out, and ways of retaining their strengths but overcoming their shortcomings are proposed. A class of networks called inference networks is introduced in order to demonstrate that logical reasoning capability can also be modeled using networks. The class of multi layer perceptrons... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Modeling robust computations in a data-flow-like style

    Publication Year: 1988, Page(s):252 - 261
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (488 KB)

    The outline of a programming language which can be used to avoid limitations and satisfy the requirements of visibility of the application and high mapping to the physical support is presented. Its expressive power is shown through a set of examples, characteristics of robustness of the applications, as well as the possibility of describing fault-tolerant applications, are discussed and exemplifie... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Domains: a new approach to distributed system management

    Publication Year: 1988, Page(s):154 - 163
    Cited by:  Papers (3)  |  Patents (8)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (676 KB)

    The domain model allows administrators to specify and structure management policies. In doing so, it provides a uniform interface across different functions of management, while being able to scale and handle multiple-organization systems. The need for subdomains and touching domains to provide the flexibility of structuring management necessary to meet the requirements of interorganizational larg... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The effective implementation of a flexible communication mechanism for distributed and parallel programming

    Publication Year: 1988, Page(s):225 - 231
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (468 KB)

    A mechanism that relies on shared communication resources but avoids many of its known disadvantages is proposed. Its main idea is introduced as an extension of the Communicating Sequential Processes (CSP) notation to asynchronous and type-safe communication behavior. It is shown that this mechanism can be used for strongly and loosely coupled processors and is thus independent of a specific distr... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Design and analysis of heuristic functions for static task distribution

    Publication Year: 1988, Page(s):124 - 131
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (584 KB)

    Two static task-scheduling algorithms are introduced. The first algorithm, called heavy-node first (HNF), is very simple and is based on a local analysis of program graph nodes at each level. The second method, called Weighted Length (WL), considers a more global view of the program graph, taking into account the relationship among the nodes at different levels. The two schemes are compared agains... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Global stability detection in the asynchronous distributed computations

    Publication Year: 1988, Page(s):87 - 92
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (420 KB)

    A simple algorithm for detecting global stability in asynchronous distributed computations is presented. it is based on observation of the outgoing channels, using a single token which acts as observer. No vector is necessary (in the transmission of the token on the ring), and this algorithm does not require the first-in/first-out (FIFO) property for the communication channels View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Network management in a large OSI environment

    Publication Year: 1988, Page(s):9 - 13
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (336 KB)

    Network management problems in a large open systems interconnection environment are addressed. Problems of integration and dividing management responsibility for a hypothetical management system are discussed. Suggestions for future research and standardization work are offered View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The recovery problem in distributed systems

    Publication Year: 1988, Page(s):357 - 365
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (544 KB)

    It is argued that the conventional transaction concept is viable in a general distributed system only with respect to its all-or-nothing property. In developing a recovery strategy, methods for damage assessment also have to be considered. This means more overhead, especially in the recovery phase, but the penalties for normal, i.e. error-free processing, can be alleviated. The ultimate goal shoul... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Transaction failure tolerance without loss of concurrency

    Publication Year: 1988, Page(s):502 - 511
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (676 KB)

    Two recovery algorithms are described: one to achieve failure tolerance of sequential transactions that read or write object, and the other to achieve failure tolerance of sequential transactions that manipulate typed objects and act on the outside world. Both require the two-phase transaction model and do not interfere with concurrency. It could be argued that the second one induces overheads in ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Computer architectures: a case for top-down design

    Publication Year: 1988, Page(s):445 - 451
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (460 KB)

    It is shown how top-down design, decomposition, abstraction, rapid prototyping and other methods from software engineering can be adapted to computer architecture design. A sample architecture design process is used to demonstrate the use of the methods. The case study deals with the upper architecture levels View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Virtual processing in ParMod

    Publication Year: 1988, Page(s):245 - 251
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (420 KB)

    It is shown that scheduling is necessary for implementing programming languages which allow unbounded parallelism. On the other hand, programming of scheduling mechanisms and policies for parallel activities is desired in the parallel-programming language itself. ParMod supports user scheduling by means of suspension, region statements, sequence parts and pre parts. User scheduling restricts syste... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Domain based configuration and name management for distributed systems

    Publication Year: 1988, Page(s):147 - 153
    Cited by:  Papers (2)  |  Patents (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (648 KB)

    An overview is given of the conic configuration environment, which applies the domain concept to the structuring of configuration and name management in a distributed system. The use of access-control lists as a security mechanism is shown. Inheritance rules are specified for access permission and access rights to maintain subdomains and overlapping domain relationships. Servers are required to ma... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Design aspects of coherent optical networks

    Publication Year: 1988, Page(s):57 - 64
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (580 KB)

    Aspects and functions of an integrated-service multichannel network are analyzed. The interaction between technological constraints and network architecture is discussed. A management scheme fully compatible with open systems interconnection is presented. This scheme shows the increased flexibility of a multichannel integrated network, which is derived from the possibility of reconfiguring the net... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • On basic rewriting rules to verify communicating machines

    Publication Year: 1988, Page(s):216 - 224
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (504 KB)

    An approach to proving some properties of the communication of communicating machines is presented. The methodology is based on a simple algebra, using rewriting systems on communicating words and homogeneous expressions. The semicommutations and partial commutations over an alphabet of sending and receiving messages suffice to guarantee the progress properties or the communication. Basic rules of... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • SALOME: a transputer-based software in a workstation dedicated to biotechnology

    Publication Year: 1988, Page(s):119 - 123
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (368 KB)

    The application of a transputer-based array to scientific calculations in the field of biotechnology is described. Energy minimization programs are examined to show the power than can be expected from parallel programming compared to conventional sequential programs in domains that are known to be very greedy in CPU time consumption. The implementation and benchmarking of the program, written enti... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The architecture of a distributed computer system

    Publication Year: 1988, Page(s):383 - 390
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (608 KB)

    Some design and implementation issues of the Ogham kernel for distributed operating systems are presented. Ogham is based on an object model, with all objects accessed in a location-independent fashion using a specialized communication facility. Resource managers are realized as processes that are free to migrate throughout the system. Ogham is designed to allow local-area-interconnected microcomp... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Assigning distinct identities to sites on an anonymous distributed system

    Publication Year: 1988, Page(s):82 - 86
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (304 KB)

    A distributed algorithm that assigns distinct and ordered identities to the sites of a connect anonymous distributed system is presented. The basis for this algorithm is the sequential distributed traversal scheme, the sequential aspect of which ensures that all sites are systematically visited in mutual exclusion. Such an algorithm allows the definition of sites to be independent of particular of... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A specification of a distributed heterogeneous systems administration

    Publication Year: 1988, Page(s):45 - 50
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (412 KB)

    Research on telecommunications networks management in a heterogeneous context is presented. This research is a part of the IDEA project (intelligence, diagnostic, expertise, administration). The purpose of IDEA is to propose a methodology for translating the manufacturers' managements to a unique user-access point, with unified syntax and semantics. In this context, a management architecture and f... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Design of an object-oriented distributed system: experience with CSA

    Publication Year: 1988, Page(s):192 - 201
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (972 KB)

    The Communication System Architecture (CSA), a five-year project to design and implement an object-oriented architecture for distributed office systems is described. The architecture is based on an object model supported by a hierarchy of recursive abstract object machines, allowing transparent communication and providing a set of facilities for building high-quality distributed applications in a ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.