Showing papers on "Shared resource published in 1990"

Strategy, Control Systems, and Resource Sharing: Effects on Business-Unit Performance

Abstract: Focusing on business units (SBUs) in diversified firms, this study investigated the relationships among control systems, resource sharing, and competitive strategies and their interactive effects o...

Multiprocessor system having statically determining resource allocation schedule at compile time and the using of static schedule with processor signals to control the execution time dynamically

Abstract: A parallel processing architecture centralized controller for containing a granting access to a shared resource for a plurality of processors by use of a transaction schedule. The controller includes a memory for storing a schedule, a pointer into the memory for accessing a particular entry in the schedule, a control circuit for asserting a grant signal to a particular one processor according to the entry in the schedule which the pointer identifies, and an incrementer for advancing the pointer to the next schedule entry upon an assertion of a release signal from the processor granted access to a shared resource. A particular processor accesses the shared resource only when its grant signal is asserted, therefore the controller imposes a particular order on the transaction of the plurality of processors and allows some execution time variations from those execution times which a compiler used in establishing the schedule. The parallel processing architecture includes a shared resource coupling the processor to allow transactions among the processors. Variations in the controller allow monitoring of the shared resource to determine decision paths and to use appropriate portions of the transaction schedule depending upon a particular decision path.

Design considerations for usage accounting and feedback in internetworks

Abstract: This paper investigates the design of resource usage feedback mechanisms for packet switched internetworks. After a discussion of the motivations for feedback mechanisms, feedback channels and policies are described. We then outline issues raised by the design of mechanisms to realize these policies, including: network service disciplines, accounting granularity, metrics, authentication, and coordination among transit carriers.Usage-based charging is only one means of feedback. Our purpose is to begin a systematic discussion of the technical issues associated with a range of usage feedback alternatives. Therefore the paper should not be read as a policy statement promoting usage sensitive charging in internets. In fact, one of the goals of the feedback mechanisms explored in this paper is to allow network service providers and users to avoid the introduction of usage sensitive charges if they so wish; while still realizing the benefits of statistical resource sharing offered by packet switching and the benefits of efficient resource utilization offered by usage feedback.

Apparatus and method for distributed dynamic priority arbitration for access to a shared resource

Abstract: In a data processing system in which resource units are shared by a plurality of processing units, an arbitration unit is disclosed wherein the priority assigned to each processing unit is dynamically assigned to equalize accessibility to the shared resource. A signal path, associated with each possible level of priority, is coupled to each processor unit. The processor unit applies an activation signal to the signal line associated with the priority of the processing unit when the processing unit has a requirement for the shared resource and an arbitration is being performed to determine access to the resource. During the arbitration procedure, each processing unit requiring access to the shared resource compares the current priority of the associated processing unit to the activation signals on the signal paths to determine when the processing unit can gain access to the shared resource. After the arbitration procedure, the processing unit priority level is redetermined by each processing unit based on a comparison of the current priority level and the highest priority level active during arbitration.

Analysis of resource sharing in information providing services

Abstract: The key to an economical service is in the sharing of physical resources among the customers. This applies to the information storage and sourcing devices, as well as to the transmission bandwidth. One of the possible solutions to the resource sharing problem is the grouping of customers with similar service requests, and broadcasting the information rather than granting the service individually. The customers' behavior (in particular, the customers' patience) is modeled, and the service performance for different service scenarios is analyzed. It is shown that in case of the threshold-activated service scenario, it is possible to optimize the service performance by using the statistics of the service requests and customer patience. >

Portable, resource sharing file server using co-routines

Abstract: Disclosed is a resource sharing system having a host computer and a plurality of a client computers connected by a Local Area Network (LAN). Resource sharing software, located in the host, is comprised of a supervisor process, a spooler process, and a plurality of session worker processes. This supervisor receives an initial call request from a client, and creates a session worker to handle all subsequent requests from that particular client. Each session worker creates a plurality of co-routines, one for each concurrent request, which operate to provide a high-performance response to the network. Co-routines share the session worker stack, as well as all other areas in the session worker. Because co-routines never interrupt each other, there is no competition for resources, and no need for critical sections of software.

Integration of CAD/CAE with multimedia teleconferencing and messaging via broadband networks and shared resource servers

Abstract: It is noted that, if multimedia electronic meeting and messaging systems were tightly integrated with networked CAE/CAD (computer-aided engineering and design) resources, great value could be added to modern design projects. Work in Bellcore's Integrated Media Architecture Laboratory (IMAL) relevant to these and other related capabilities is described. A working premises-based network with shared CAD/CAE systems, conferencing, and messaging servers, encompassing video, graphics, text and audio, has been constructed as part of the Bellcore IMAL project. The example IMAL network can be duplicated with off-the-shelf products and can be extended to link multiple premise locations through the use of commonly available DS-3 codecs and telephone-company-provided DS-3 fibers. >

Retrospective on DACNOS

Abstract: Heterogeneity of hardware and software is a fact in most distributed computing environments. The DACNOS prototype is a network operating system that enables resource sharing in such environments. It extends the local operating systems without interfering with existing programs. It provides comprehensive system level support for distributed applications.

Distributed management of replicated and partitioned files under DRAGON SLAYER

Abstract: The authors describe the file management system of DRAGON SLAYER, a distributed operating system which provides transparent resource sharing in a heterogeneous local area network with completely decentralized control. Users are able to use any of the system files for which they have authorization, without regard to the individual characteristics of the computers on which the files reside or the physical locations of the users and files. In addition, files can be dynamically fragmented, and the fragments can be replicated and relocated dynamically. Copies are kept mutually consistent while all file operations work either transparently on the whole file or on specific fragments. The DRAGON SLAYER resource management supporting these file system operations is presented. Beyond the obvious advantages of file access reliability and flexibility, the authors present simulation results proving the superior performance of this file system as compared with conventional distributed file systems with non-fragmented files. >

Access control in a workstation-based distributed computing environment

Abstract: The Institutional File System (IFS) project is a joint project of the Center for Information Technology Integration at the University of Michigan and the IBM Corporation. The goal of the IFS project is to offer network file services that are institution-wide, transparent, reliable, and secure. The mechanisms used to control access to system services on the IFS project are described. A logical overview of access control at the IFS project is given and is followed by an enumeration of the services on which the project relies, their interdependencies, and the mechanisms by which access to these services is secured. Present vulnerabilities and future plans are described. >

Agricultural libraries and information centres in China: cooperation, resource-sharing and networking

Abstract: This research is aimed towards the design and planning of an agricultural library and information network in China. Systems approach is applied to the construction of descriptive, mathematical and hybrid (logical) models in the design phase of the system development cycle. Four major conventional network functions are chosen as core functions to be designed in detail. A number of procedures and issues are identified for both overall network and individual function design. The embodiment of the design is based on the particular context to which it relates. The results of the design arc presented by either descriptive, or mathematical or hybrid (logical) models, decided by the categories of issues (quantitative or qualitative, or hybrid). The impacts of new technologies are explored and three alternative programmes, dependent on the technologies to be applied, arc defined. The cost-effectiveness-benefits of the services as overall criteria, and AHP (Analytical Hierarchy Process) as an processing tool, help determine 1) which program will be run, computerized or non-computerized (by cost-benefit analysis}; and 2) for a particular program, the resource allocations (mainly budgets) among co-operative activities and the resource allocations within the activity, i.e. two-level allocations. Zhejiang province, one of 29 provinces in the country, is taken as an example of regional (provincial) network model. Twenty-eight nodes Iron: five sectors and three levels, plus more potential nodes, will be configured hierarchically within the respective sectors and, in decentralised mode among sectors, in terms of network management. But, in terms of information flow and transaction, the configuration will be a mixed one. The purposes of data collection arc identified at four stages of network planning, i.e. investigation of situation, theory-testing, explanation of model and prediction. A number of investigations were made to fulfil the objectives of data collection. The descriptions of those investigations are given and the results arc discussed. The barriers to data collection confronted in Chinese circumstances are presented. Statistical analyses are carried out for certain types of data 1) to seek the conformity with the empirical assumptions, 2) to help set objective measures, 3) to estimate parameters and co-efficiency, and 4) to derive some mean, average and unit values. The research has focused on the organisational aspects of networking. However, the macro-consideration of technology is inevitable since it is a time of technological change. The latter stages of mathematical modelling, i.e. optimisation, prediction and validation are left until more empirical data become available. Nevertheless, it is possible for the author to make recommendations about the development of agricultural library and information networks in China.

Asynchronous Circuits for Token-Ring Mutual Exclusion

Abstract: : We have described three algorithms for distributed mutual exclusion on a ring. All algorithms use a token to select a candidate. We have already implemented the most efficient of these algorithms as an asynchronous VLSI circuit. We are now going to implement the simplest one. An arbitrary number (> 1) of cyclic automata, called "masters," make independent requests for exclusive access to a shared resource. The circuit should handle the requests from the masters in such a way that any request is eventually granted, and there is at most one master using the shared resource at any time. The masters are independent of each other: They do not communicate with each other, and the activity of a master not using the resource should not influence the activity of other masters. A master, M, communicates with its private server, m. When M wants to use the shared resource (M is said to be a candidate), it issues a request to m. When the request is accepted, M uses that resource (for a finite period of time), and then informs m that the resource is free again. The servers are connected in a ring. At any time, exactly one (arbitrary) server holds a "privilege," or "token." The token circulates continuously around the ring of servers, and only the server that holds the token may grant the resource to its master, which guarantees mutual exclusion on the access to the resource.

Resource binding—a universal approach to parallel programming

Abstract: The suitability of a parallel programming paradigm on a particular parallel architecture often greatly influences the overall system performance. This report presents the parallel programming paradigm, resource binding, which offers an architecture independent environment for various parallel computation models. The resource binding technique, with a simple set of primitives, enables a flexible and consistent way of handling process synchronization, shared resource management, and other problems in parallel programming. High portability can be maintained without losing performance. For a clear introduction to the technique, comparisons are made among resource binding and other well-defined schemes such as semaphore, monitor, message passing and Linda.

Resource sharing: practice and problems

Abstract: Resource sharing is an important element in the national planning of library and information services to meet the needs of information, education and culture of the whole community at all levels. An overview of resource sharing practices is presented, with particular reference to the British scene. It is also argued that, with the approach of the Single Market in 1992, resource sharing should now be considered on a European scale. In conclusion, some problems associated with the practice of resource sharing are considered.

Using right control system

Abstract: PURPOSE:To speed up a processing guiding the list of resources which a certain user can use by storing using right information in a storage device for respective users and retrieving using right information on the users corresponding to inputted user identification names at a high speed. CONSTITUTION:A using right display means 101 starts a using right file selection means 103 by using the inputted user identification names, and therefore the using right file name of a using right file 105 and the shared resource using right file name of a shared resource using right file 106 are obtained. Then, using right informations stored in the using right file 105 and the shared resource using right file 106 are obtained by starting a using right retrieval means 104 by using the obtained file name. The resources which the user can use and the using right are displayed by starting a display means 102 by using obtained using right information. Thus, the processing guiding the list of resources which a certain user can use can be performed at a high speed.

Advances in library resource sharing

Abstract: Published annually, this third volume includes a discussion of access versus warehousing, resource sharing versus collection building, articles on the history of resource sharing, traditional means of resource sharing and case studies. A final section deals with the future of resource sharing and focuses on technological implicatinos, the impact of CD-ROM on resource sharing, document delivery service alternatives, fee versus free and electronic publishing.

Boxes and boards for sharing resources

Abstract: Computer users usually type at their own personal computer keyboards and look at their own display screens but, on occasion, every user needs to access a printer, or some other resource, even another computer

Intelligent gateways for unified computing environments

Abstract: The use of gateways to integrate heterogeneous networks is investigated. The goal is to extend the unified environments across network boundaries transparently so that services provided by one environment are available to other networks. This approach is most effective if the connected networks have incompatible environments which provide similar services. Intelligent gateways have the advantages of low cost, modularity, resource sharing across the networks and minimal modification to the existing systems. The basic concept of an intelligent gateway is introduced. A case study in which an intelligent gateway supports a network file system between a SUN network and a PC network is presented to demonstrate the basic ideas. >

Blocking for Single-Stage Resource Sharing

Abstract: The design and management of communication networks involve the allocation of network resources, especially the capacity of transmission links, for transporting the information generated by communication terminals. This chapter focuses on the sharing of transmission links, whereas the same theory can be applied to other problems, such as the sharing of memory and processors.

Resource binding-a universal approach to parallel programming

Abstract: The authors present a parallel programming paradigm, resource binding, which offers an architecture-independent environment for various parallel computation models. The resource binding technique, with a simple set of primitives, enables a flexible and consistent way of handling process synchronization, shared resource management, and other problems in parallel programming. High portability can be maintained without losing performance. For a clear introduction to the technique, comparisons are made among resource binding and other well-defined schemes such as semaphore, monitor, message passing, and Linda.

Efficient caching of temporary files

Abstract: Caching has emerged as an important method of improving the performance of distributed file systems. May researchers have shown that network traffic, server load and file access times can all be significantly reduced by caching files on the client. The authors propose to improve cache performance by increasing flush times for temporary files and performing directory operations of temporary files on the client since temporary files have unique features that allow it. >