Showing papers on "Shared resource published in 1991"

Resource allocation in a dynamically partitionable bus network using a graph coloring algorithm

Abstract: An efficient dynamic graph traversal algorithm is used to identify nonconflicting requests and to allocate network resources in a dynamically partitionable bus network (DPBN). In centralized network control a special processor receives from the control computer of a partitionable bus network an adjacency matrix which indicates conflicts among requests. It applies the dynamic graph traversal algorithm and returns the identified nonconflicting requests to the control computer. The control computer then physically partitions the network into a number of subnetworks for processing the nonconflicting requests in parallel. In distributed control, each station determines conflicts and sets the switches. The results of performance evaluation show a 40% decrease of network delay as compared with a fully utilized, but unpartitioned local area network. >

Shared computer resource allocation system having apparatus for informing a requesting computer of the identity and busy/idle status of shared resources by command code

Abstract: Computer system resources shared by several central processing units are allocated by allowing one processing unit to temporarily gain exclusive access to a particular shared resource. Access to a particular resource is controlled by a memory location which contains information representing the current state of the resource and the identity of any processing element currently utilizing the resource. In the case where several resources are interchangeable, the memory location may also contain information regarding the busy/idle states of other interchangeable resources. The memory location can be interrogated by any of the processing elements via command and address information. If the contents of the memory location indicate that the associated resource is not in use, then the interrogating processing element immediately obtains control of the resource. If the resource is in use, the identity of the processing element currently using the resource and any stored resource state information is returned to the interrogating processing element which can then continue processing without the resource or choose another idle resource.

A priority ceiling protocol for multiple-instance resources

Abstract: A systematic optimization process for multiple-instance resource sharing in real-time systems is presented. The authors derive the schedulability condition for systems with such resources and present an algorithm which can be used to divide a resource pool into smaller groups in order to improve the worst-case blocking behavior. They present the system model used and review some related work. The multi-instance priority ceiling protocol and its properties are discussed, the effect of resource preallocation on the schedulability of a system is studied, and an optimal resource preallocation algorithm is presented. >

Asynchronous arbiter state machine for arbitrating between operating devices requesting access to a shared resource

Abstract: A system has a shared resource, such as a bus or memory, with which various devices may communicate upon a request being granted by an arbiter. In order to reduce the arbitration time, the arbiter is comprised of a state machine and latch which run asynchronously. Consequently, once one request has been granted and acted upon, the state machine will commence arbitration for any remaining requests.

Multiprocessor shared resource management system implemented as a virtual task in one of the processors

Abstract: A shared resource management system in a multi-processor system comprises the steps of: requesting an accessing right from a processor requesting an accessing right to a shared resource to one processor in the system; taking the place of the request for the accessing right to the shared resource by the one processor; and establishing the accessing right by the one processor when the accessing is permitted and informing the acquirement of the accessing right to the accessing right requesting processor.

Contracts in Database Federations

Abstract: A model for information sharing in a loose association of information/database systems is proposed Such a system is referred to as a federation, where the fundamental characteristic is the autonomy of the member systems (the nodes) The model for sharing is based on the establishment of bilateral agreements between nodes These agreements are represented by contracts, the terms of which define the rules for information exchange The model provides flexibility for resource sharing and for cooperation in a decentralized setting

Supporting real-time concurrency

Abstract: Concurrent real-time applications are complicated since both timing and consistency constraints must be met for correct performance. Furthermore, techniques to enforce these two forms of constraints are often incompatible. For instance, priority-driven preemptive scheduling, which is optimal for meeting timing constraints in some systems, may leave a shared resource's state inconsistent. On the other hand, mutual exclusion techniques that ensure the consistency of shared resources are not well-suited to meeting timing constraints. This dissertation develops concepts and programming language constructs for facilitating the enforcement of both real-time and consistency constraints in applications with concurrency. Our programming paradigm combines an object-based paradigm for the specification of shared resources, and a distributed transaction-based paradigm for the specification of application processes. Resources provide abstract views of shared system entities, such as devices and data structures. Each resource has a state and defines a set of actions that can be invoked by processes to examine or change its state. A resource also specifies scheduling constraints on the execution of its actions to ensure the maintenance of its state's consistency. Processes access resources by invoking actions and express precedence, consistency and timing constraints on action invocations. The implementation of our language constructs with real-time scheduling and locking for concurrency control is also described, including a novel deadlock prevention technique. The utility of the constructs are demonstrated in two ways. First, we describe their use to solve a general concurrent real-time problem called timed atomic commitment. We then describe how they were used to program a graphic simulation of two robot arms coordinating to pick up a moving object under timing constraints.

Specification and design of shared resource arbitration

Abstract: The specification, modular design and verification of distributed communicating systems is demonstrated by an example. The scheduling of the access to a common shared resource by a finite number of devices with priorities is a well known problem from hardware and operating systems design. Using the concepts of stream processing continuous function a variety of formal requirement and design specifications for this problem are given and the design specifications are proven correct w.r.t the requirement specifications. From the design specifications one can immediately read off applicative programs implementing the requirement specifications. Special attention is paid to the aspect of modelling time.

Exclusive control system

Abstract: PURPOSE:To realize exclusive control allowing one or more transaction processing programs constituting the same on-line processing to occupy a shared resource requiring competition management. CONSTITUTION:This exclusive control system is provided with an occupation management table 9 registering occupation information for identifying on-line processing capable of occupying a shared file 8, an occupying means 5 for informing the permission of occupation to a processing program generating a shared file occupation request out of plural transaction processing programs 3-1 to 3-n at the time of judging that the program is a program capable of occupying the file 8 while referring the table 9, and an occupation resetting means 6 for informing the permission of occupation reset to the transaction processing program generating the a shared file occupation reset request out of the programs 3-1 to 3-n at the time of judging that the program is a processing program constituting the on-line processing occupying the file 8 while referring the table 9.

Benchmarking a network storage service

Abstract: Benchmarking a network file server introduces some unique considerations over traditional benchmarking scenarios Since the user is executing on a client system interconnected to the file server, the client and network must be provided for during benchmarking During a recent procurement action, Sandia National Laboratories (SNL) was challenged to develop a benchmark suite that would accurately test the network requirements The authors describe the benchmark design and summarize the experience gained from the benchmark execution SNL offered three possible benchmark configurations All options were used by different vendors Therefore, each option was tested and each successfully executed the benchmark tests Since the tests employed the actual commands the end users will execute, SNL feels that it has obtained a high level of assurance that end-user functionality and performance have been achieved >

Shared resource access controller

Abstract: PURPOSE:To suppress the reduction of throughput for an entire device caused by increasing the number of processors by connecting respective shared resources through a bus controller by a shared bus and providing a bus arbiter at the shared bus so as to arbitrate shared bus access requests to be outputted from the respective processors CONSTITUTION:Shared resources 1,100, 1,101 equipped with the same contents are provided for respective processors 1,000, 1,001 and at the write cycle time of any one processor, the contents of the respective shared resources are simultaneously changed through a shared bus 1,310l so that the contents of the respective shared resources can be always maintained same Thus, since the respectively correspondent shared resources are accessed at the time of a read cycle (data fetch cycle) not to change data on the shared resources by the respective processors, access competition is not generated onto the shared resources and further, since the contents of the shared resources provided corresponding to the respective processors are simultaneously changed through the shared bus at the time of the write cycle, the throughput is prevented from being lowered

Computer Aided Storage Design

Abstract: Several alternatives are available to solve the lack of space at libraries nation-wide: weeding, reformatting, resource sharing or building/remodeling collection storage facilities. The author describes a computer-aided storage design method for compact shelving using dBase 111 + software.

CheapLANS: Resource Sharing on a Budget.

Abstract: Presentation et evaluation de quelques solutions de problemes d'incompatibilite entre microordinateurs et d'options economiques de transfert electronique d'information (partage de fichiers, d'imprimantes, de fichiers et d'imprimantes, progiciels de connection)

Evolutionary steps toward a distributed operating system: theory and implementation

Abstract: Generally speaking, distributed operating system (DOS) designers seem more concerned with resource sharing, global file system transparency, and implementation methodology. We feel that a DOS may provide the platform for linear speedup of applications if performance considerations are given highest priority in the design of the DOS. Work has been underway to test the implementation of such a DOS design in the MINIX environment.

Multiprocessor shared resource control system

Abstract: PURPOSE:To prevent a shared resource from being abnormally occupied by providing the first means to show whether the shared resource is occupied by any one of processors or not, and the second means to be turned to a certain state different from any previous states each time the processor to occupy the shared resource is changed. CONSTITUTION:When using a shared resource 4, a processor 1 investigates a lock flip-flop 41 of the shared resource 4 and when lock is enabled, a lock counter 42 is incremented. Then, a processing is exerted onto the occupied shared resource 4 and the lock is released. When the lock is disabled since the other processor, for example, a processor 2 already occupies the shared resource 4, the value of a lock counter 2 is read. After starting the possession of the lock, when there is no change in the value of the lock counter 2 even after the lapse of the maximum resource occupation time, the lock counter 42 is incremented and the lock is released.

Operating systems for real-time applications

Abstract: In the world of large computers, operating systems (OSs) have been with us for quite some time. In fact the elementary ones go back to the 1950s. Major steps were made in the 1960s, and by the mid 1970s their concepts, structures, functions and interfaces were well established.

Resource sharing: priorities, conflicts and rising expectations

Abstract: The purpose of this paper is to describe in general the economic realities of resource sharing, the place of resource sharing in the scholarly information structure, and the con flicts which arise in setting priorities for our collections and services.1 Resource sharing has many components, including interlibrary loan and reciprocal access and borrowing ar rangements.2 Conflicts can arise from the ownership versus access debate, the cost of access, inadequate bibliographic control, and inconsistency of supply. Discussed here will be interlibrary loan in particular with emphasis on the issues

An outline for planning an academic wide-area network

Abstract: For those unfamiliar with the concept of a wide-area network (WAN) the following definition will help to orient the reader. A wide-area network is a concept of configuration and integration of resources designed to complement a site's current LAN and mainframe technologies while providing a coherent and organized approach to the integration of personal computer-based network technology to a wider geographical region. Wide-area networking involves a diversity of technologies that cover broad areas of functionality, design and implementation. A wide-area network is, in each instance, unique; it is predicated upon unique needs and constraints for a given installation. A WAN can provide educational institutions with: * Connectivity; * Resource sharing; * Off-loading mainframe processing; * Integration of mainframe co-processing; and * Efficient and effective e-mail systems. Implementation Guidelines Following are several guidelines that should be considered when planning a wide-area network implementation. I. Intended Utilization A wide-area network may be utilized for a diversity of applications. Based upon the intended use, a WAN's design characteristics may encompass a wide latitude of technologies. A good design will proceed from the top down, since the user's needs dictate much of the technology required. Ia. Demands of academic wide-area networking can necessitate a design that will exceed those of a commercial user. Ib. An accurate evaluation of WAN utilization is mandatory. Without a thorough review of initial and potential use of the network, its design would be predicated on potentially false assumptions. This could evolve into a design that is either deficient in performance and flexibility or is excessive and incompatible with the organization's currently established base of installed technology. II. Diversity of Technologies Wide-area technology encompasses a broad variety of available components. These components cover the gamut in performance capabilities. Integration of these diverse technologies offers significant challenges to the WAN designer. III. Incorporate Appropriate Technologies Incorporating state-of-the-art technology into a wide-area network can be quite challenging. Allowing for future expansion and future utilization are extremely important issues in the design of an academic WAN. It must incorporate maximum flexibility and effectively utilize the available technologies. IIIa. Supporting academic programs through a wide-area network could easily require higher levels of support than those for the business environment. This is due to characteristics of academic computing such as constant student and staff turnover (they graduate or move), users' levels of expertise ranging from nil to infinity, and the ever-changing nature of semester-based coursework (one semester requires statistical software, the next needs graphics). IIIb. In many instances the performance characteristics needed for an academic WAN will exceed those for a business wide-area network. At colleges and universities, for example, research and advanced applications demand optimum performance from hardware. In addition, the amount of data traffic can be phenomenal, depending on the size and nature of the school. IIIc. Ease of operation, minimal maintenance and serviceability are key considerations for an academic wide-area network. IV. Testing and Evaluation Develop adequate evaluation strategies in cooperation with both the hardware manufacturers and the WAN designers. Utilize design goals to accurately measure performance over a broad range of conditions. Test key components and the initial design's characteristics thoroughly. Develop performance standards for peripherals. In other words, push the technology to its limits. …

Composite electronic computer

Abstract: PURPOSE: To prevent the unnecessary use of computer resources to perform the exclusive control of high reliability among plural computers. CONSTITUTION: The exclusive control using a semaphore mechanism 111 of an electronic computer 11 is executed, and P and V operation instructions to the semaphore mechanism 111 are issued in accordance with a use request and a use end report of a shared resource 14 from another computer 12, 13 or the computer 11 itself. It is determined whether permission for the use request should be reported to the use request source device or not in accordance with the value of a semaphore variable. Therefore, permission is not returned to another computer 12 or 13 or the specific computer 11 merely and a lock server mechanism 112 is not set to the execution wait state by a failure of the P operation instruction even if the use of the shared resource 14 is not permitted. COPYRIGHT: (C)1992,JPO&Japio

Exclusive control system for resource

Abstract: PURPOSE:To occupy a resource over plural tasks by providing a source registration table, which registers the name of the resource to be dynamically generated by a use declaration from the task, to be chained to a resource sharing group control table and deleting the resource registration table when there is no shared task. CONSTITUTION:A resource sharing group control table 10 controls names 11 of task groups to share the resource and used for deciding the task knowing the task group name as the task group which can share the resource. A resource registration table 20 is dynamically generated by the use declaration of the task and manages a name 22 of the resource to be shared. Thus, the resource to be dynamically shared by the plural tasks can be registered.

Naming in the distributed operating system ZGL

Abstract: In this paper, the naming scheme used in the heterogeneous distributed operating system ZGL is described and some of the representative techniques utilized in current distributed operating systems are examined. It is believed that the partitioning of the name space into many local name spaces and one global shared name space allows the ZGL system to satisfy each workstation's demand for local autonomy and still be able to facilitate transparent resource sharing. By the division of the system into clusters and the use of a combined centralized distributed naming mechanism, the system is able to avoid both the bottleneck problem caused by a single centralized name server for the whole system and the per formance degradation due to a full distributed scheme.