Showing papers on "Shared resource published in 1998"

A Quantitative Measure Of Fairness And Discrimination For Resource Allocation In Shared Computer Systems

Abstract: Fairness is an important performance criterion in all resource allocation schemes, including those in distributed computer systems However, it is often specified only qualitatively The quantitative measures proposed in the literature are either too specific to a particular application, or suffer from some undesirable characteristics In this paper, we have introduced a quantitative measure called Indiex of FRairness The index is applicable to any resource sharing or allocation problem It is independent of the amount of the resource The fairness index always lies between 0 and 1 This boundedness aids intuitive understanding of the fairness index For example, a distribution algorithm with a fairness of 010 means that it is unfair to 90% of the users Also, the discrimination index can be defined as 1 - fairness index

A Quantitative Measure Of Fairness And Discrimination For Resource Allocation In Shared Computer Systems

Abstract: Fairness is an important performance criterion in all resource allocation schemes, including those in distributed computer systems. However, it is often specified only qualitatively. The quantitative measures proposed in the literature are either too specific to a particular application, or suffer from some undesirable characteristics. In this paper, we have introduced a quantitative measure called Indiex of FRairness. The index is applicable to any resource sharing or allocation problem. It is independent of the amount of the resource. The fairness index always lies between 0 and 1. This boundedness aids intuitive understanding of the fairness index. For example, a distribution algorithm with a fairness of 0.10 means that it is unfair to 90% of the users. Also, the discrimination index can be defined as 1 - fairness index.

Real time backup system

Abstract: In a computer network system, a user-defined file modification request is communicated to a primary server, which communicates the request to a secondary server. The file modification request is saved in a non-volatile storage media associated with the primary server, and the file modification request is executed and saved in a non-volatile storage media associated with the secondary server.

Virtual machine monitors for scalable multiprocessors

Abstract: The problem of extending modern operating systems to run efficiently on large-scale shared memory multiprocessors without a large implementation effort is solved by a unique type of virtual machine monitor. Virtual machines are used to run multiple commodity operating systems on a scalable multiprocessor. To reduce the memory overheads associated with running multiple operating systems, virtual machines transparently share major data structures such as the operating system code and the file system buffer cache. We use the distributed system support of modem operating systems to export a partial single system image to the users. Two techniques, copy-on-write disks and the use of a special network device, enable transparent resource sharing without requiring the cooperation of the operating systems. This solution addresses many of the challenges facing the system software for these machines. The overheads of the monitor are small and the approach provides scalability as well as the ability to deal with the non-uniform memory access time of these systems. The overall solution achieves most of the benefits of operating systems customized for scalable multiprocessors yet it can be achieved with a significantly smaller implementation effort.

Software partitioned multi-processor system with flexible resource sharing levels

Abstract: Multiple instances of operating systems execute cooperatively in a single multiprocessor computer wherein all processors and resources are electrically connected together. The single physical machine with multiple physical processors and resources is subdivided by software into multiple partitions, each running a distinct copy, or instance, of an operating system. Each of the partitions has access to its own physical resources plus resources designated as shared. The partitioning is performed by assigning all resources with a configuration tree. None, some, or all, resources may be designated as shared among multiple partitions. Each individual operating instance will generally be assigned the resources it needs to execute independently and these resources will be designated as “private.” Other resources, particularly memory, can be assigned to more than one instance and shared. Shared memory is cache coherent so that instances may be tightly coupled, and may share resources that are normally allocated to a single instance. This allows previously distributed user or operating system applications which usually must pass messages via an external interconnect to operate cooperatively in the shared memory without the need for either an external interconnect or message passing. Examples of application that could take advantage of this capability include distributed lock managers and cluster interconnects. Newly-added resources, such as CPUs and memory, can be dynamically assigned to different partitions and used by instances of operating systems running within the machine by modifying the configuration.

Data partitioning and load balancing in parallel disk systems

Abstract: Parallel disk systems provide opportunities for exploiting I/O parallelism in two possible ways, namely via inter-request and intra-request parallelism. In this paper, we discuss the main issues in performance tuning of such systems, namely striping and load balancing, and show their relationship to response time and throughput. We outline the main components of an intelligent, self-reliant file system that aims to optimize striping by taking into account the requirements of the applications, and performs load balancing by judicious file allocation and dynamic redistributions of the data when access patterns change. Our system uses simple but effective heuristics that incur only little overhead. We present performance experiments based on synthetic workloads and real-life traces.

Clustered file management for network resources

Abstract: Methods for operating a network as a clustered file system is disclosed. The methods involve client load rebalancing, distributed Input and Output (I/O) and resource load rebalancing. Client load rebalancing refers to the ability of a client enabled with processes in accordance with the current invention to remap a path through a plurality of nodes to a resource. Distributed I/O refers to the methods on the network which provide concurrent input/output through a plurality of nodes to resources. Resource rebalancing includes remapping of pathways between nodes, e.g. servers, and resources, e.g. volumes/file systems. The network includes client nodes, server nodes and resources. Each of the resources couples to at least two of the server nodes. The method for operating comprising the acts of: redirecting an I/O request for a resource from a first server node coupled to the resource to a second server node coupled to the resource; and splitting the I/O request at the second server node into an access portion and a data transfer portion and passing the access portion to a corresponding administrative server node for the resource, and completing at the second server nodes subsequent to receipt of an access grant from the corresponding administrative server node a data transfer for the resource. In an alternate embodiment of the invention the methods may additionally include the acts of: detecting a change in an availability of the server nodes; and rebalancing the network by applying a load balancing function to the network to re-assign each of the available resources to a corresponding available administrative server node responsive to the detecting act.

QoS-aware resource management for distributed multimedia applications

Abstract: The ability of operating system and network infrastructure to provide end-to-end quality of service (QoS) guarantees in multimedia is a major acceptance factor for various distributed multimedia applications due to the temporal audio-visual and sensory information in these applications. Our constraints on the end-to-end guarantees are (1) QoS should be achieved on a general-purpose platform with a real-time extension support, and (2) QoS should be application-controllable. In order to achieve the users acceptance requirements and to satisfy our constraints on the multimedia systems, we need a QoS-compliant resource management which supports QoS negotiation, admission and reservation mechanisms in an integrated and accessible way. In this paper we present a new resource model and a time-variant QoS management, which are the major components of the QoS-compliant resource management. The resource model incorporates, the resource scheduler, and a new component, the resource broker, which provides negotiation, admission and reservation capabilities for sharing resources such as CPU, network or memory corresponding to requested QoS. The resource brokers are intermediary resource managerss when combined with the resource schedulers, they provide a more predictable and finer granularity control of resources to the applications during the end-to-end multimedia communication than what is available in current general-purpose networked systems. Furthermore, this paper presents the QoS-aware resource management model called QualMan, as a loadable middleware, its design, implementation, results, tradeoffs, and experiences. There are trade-offs when comparing our QualMan QoS-aware resource management in middleware and other QoS-supporting resource management solutions in kernel space. The advantage of QualMan is that it is flexible and scalable on a general-purpose workstation or PC. The disadvantage is the lack of very fine QoS granularity, which is only possible if supports are built inside the kernel. Our overall experience with QualMan design and experiments show that (1) the resource model in QualMan design is very scalable to different types of shared resources and platforms, and it allows a uniform view to embed the QoS inside distributed resource managements (2) the design and implementation of QualMan is easily portables (3) the good results for QoS guarantees such as jitter, synchronization skew, and end-to-end delay, can be achieved for various distributed multimedia applications.

Direct storage of recovery plan file on remote server for disaster recovery and storage management thereof

Abstract: Disclosed are a method, a storage management system, an article of manufacture comprising a computer readable medium, and a computer program product for saving a recovery plan file for a storage management server. The storage management system has a plurality of storage management servers at sites remote from one another coupled by a server-to-server infrastructure. A recovery plan file is saved for one of the storage management servers at one of the sites by establishing the server as a source for its recovery plan file. Another storage management server at a site remote from the source server site is established as a target for the recovery plan file. The source server transmits the source recovery plan file from the source server to the target server at the remote site over the server-to-server infrastructure. The source recovery plan file is managed at the target server according to defined criteria, for placement, backup, migration and expiration under the control of the source server.

Distributed System Design

Abstract: From the Publisher: This reference outlines the main motivations for building a distributed system, including inherently distributed applications, performance / cost, resource sharing, flexibility and extendibility, availability and fault tolerance, and scalability Presenting basic concepts, problems, and possible solutions, Distributed System Design serves graduate students in distributed system design as well as computer professionals analyzing and designing distributed / open / parallel systems

A resource query interface for network-aware applications

Abstract: Development of portable network-aware applications demands an interface to the network that allows an application to obtain information about its execution environment. The paper motivates and describes the design of Remos, an API that allows network-aware applications to obtain relevant information. The major challenges in defining a uniform interface are network heterogeneity, diversity in traffic requirements, variability of the information, and resource sharing in the network. Remos addresses these issues with two abstraction levels, explicit management of resource sharing, and statistical measurements. The flows abstraction captures the communication between nodes, and the topologies abstraction provides a logical view of network connectivity. Remos measurements are made at network level, and therefore information to manage sharing of resources is available. Remos is designed to deliver best effort information to applications, and it explicitly adds statistical reliability and variability measures to the core information. The paper also presents preliminary results and experience with a prototype Remos implementation for a high speed IP based network testbed.

A fully-pipelined fixed-latency communications system with a real-time dynamic bandwidth allocation

Abstract: The present invention provides for an on-chip communications method with fully distributed control combining a fully-pipelined, fixed latency, synchronous bus (35) with a two-level arbitration scheme where the first level of arbitration is a framed, time division multiplexing arbitration scheme and the second level is a fairly-allocated round-robin scheme implemented using a token passing mechanism. Both the latency and the bandwidth allocation are software programmable in real-time operation of the system. The present invention also provides for a communication system where access to a shared resource (10, 15, 20) is controlled by the communications protocol. Access to and from the shared resource (10, 15, 20) from the subsystem is through a bus interface module (40, 45, 50, 55, 60). The bus interface modules (40, 45, 50, 55, 60) provide a level of indirection between the subsystem (25, 30) to be connected to the shared resource (10, 15, 20). This allows the decoupling of system performance requirements. Communication over the bus (35) is fully memory mapped.

Heterogeneous network file access

Abstract: A method for a client to access data files residing on a first data server through a network includes coupling a heterogenous proxy server to the first data server through a first local network protocol, selectively receiving at the heterogeneous proxy server a data file from the first data servers by employing the first local network protocol, translating the data file into a format compatible with transmission through the network, and transmitting the translated data file to the client across the network.

Priority arbiter with shifting sequential priority scheme

Abstract: An improved data processing system and in particular an improved data processing system that more effectively manages a shared resource within a data processing system. More specifically, a method and apparatus for managing access to a shared resource between a plurality of devices simultaneously requesting access to the shared resource. The present invention implements a design that combines a priority configuration and a shifting sequential configuration. The access is controlled by an arbiter that determines access to the shared resource by granting first, to priority devices and then to the highest priority shifting sequential device requesting access within one clock cycle of a device terminating its request for access to the shared resource. In addition, the present invention employs a dynamic shifting sequential priority scheme by assigning lowest priority to a shifting sequential device once the device terminates its request to access the shared resource while simultaneously incrementally increasing the priority levels of the remaining shifting sequential devices.

Printer enclosing network computer and computer network system employing the same

Abstract: A subject data file to be printed out and an application corresponding to the subject data file are downloaded from a computer network to an application download type printer enclosing a network computer. Then, the application is started up on the network computer, whereby the subject data file is opened, converted to a raster image, and printed out.

Cyclic multicasting or asynchronous broadcasting of computer files

Abstract: A client/server network system is disclosed for cyclic multicasting of an image file from a central data provider (server) to one or more remote client machines (workstations) over a computer network with minimum network transmission while allowing any number of client machines (workstations) to download the image file at any moment in time without the need to synchronize with the central server's transmission. The network system includes a computer network; a plurality of remote client machines on the computer network; and a central server for providing a cyclic multicasting of an image file to one or more client machines over the computer network concurrently through the use of different transmission cycles of a single cyclic multicast session. An image file is asynchronously transferred from the central server to one or more remote client machines concurrently over a computer network through the use of different transmission cycles during a single cyclic multicast session so as to keep the network transmission and network bandwidth to a minimum.

Method for managing shared resources

Abstract: An enqueue structure type is provided within a coupling facility to record the use state and to manage a request queue of resources shared among programs on computers of a system. All programs to use a shared resource issue a request to reserve the shared resource to the coupling facility before using it. The coupling facility affords the permission to use the resource to the programs if it is usable, and adds the programs to the resource request queue if unusable. When a program using the shared resource frees it, the coupling facility fetches a program from the beginning of the request queue, reserves the shared resource for the next program and notifies the next program that the use of the shared resource has been approved. On receiving the notification, the program uses the shared resource without asking the coupling facility for the approval of the use of the shared resource again. Further, exclusive control of shared resources by use of a spin lock having shared and exclusive attributes is implemented by using two locks for one shared resource, wherein the first lock is reserved and then freed after the second lock is reserved.

Computer system, and file resources switching method applied to computer system

Abstract: A computer system comprising a server computer and a network computer operating upon downloading programs and data necessary for data processing from the server. The network computer comprises a local storage device in which at least a part of resources including the programs and data stored in a file system of the server are stored as a replica, determination means for determining a connection state to the server, and file system allocation means for selectively allocating the file system of the server and a local file system stored as a replica in the local storage device to the network computer in accordance with the determination result made by the determination means so that a common file access pass to both the file system of the server and the local file system of the local storage device can be used.

Toward a framework for power control in cellular systems

Abstract: Efficiently sharing the spectrum resource is of paramount importance in wireless communication systems, in particular in Personal Communications where large numbers of wireless subscribers are to be served. Spectrum resource sharing involves protecting other users from excessive interference as well as making receivers more tolerant to this interference. Transmitter power control techniques fall into the first category. In this paper we describe the power control problem, discuss its major factors, objective criteria, measurable information and algorithm requirements. We attempt to put the problem in a general framework and propose an evolving knowledge-bank to share, study and compare between algorithms.

Transaction locks for high availability

Abstract: One embodiment of the present invention provides a method and an apparatus that ensures proper semantics for operations when operations are restarted on a secondary server in the event of a failure of a primary server. This embodiment keeps a record on the secondary server of which operation currently has exclusive access to a shared resource. The method operates by receiving a message from the primary server indicating that a first operation on the primary server has acquired exclusive access to the shared resource. In response to this message, the system updates state information, at the secondary server, to indicate that the first operation has exclusive access to the shared resource and that any prior operations have completed their exclusive accesses to the shared resource. Upon receiving notification that the primary server has failed, the secondary server is configured to act as a new primary server. When the secondary server subsequently receives an operation retry request from a client of the primary server, it performs one of several operations. If the operation retry request is for the first operation, the system knows the first operation had exclusive access to the shared resource on the primary server. In this case, the secondary server acquires exclusive access to the shared resource, and completes the first operation. If the operation retry request is for a prior completed operation, the system returns to the client a saved result of the prior operation if the client has not received such saved result. Another embodiment of the present invention includes more that one secondary server.

Scheduling parallel applications in distributed networks

Abstract: Prophet is a run-time scheduling system designed to support the efficient execution of parallel applications written in the Mentat programming language (Grimshaw, 1993). Prior results demonstrated that SPMD applications could be scheduled automatically in an ethernet-based local-area workstation network with good performance (Weissman and Grimshaw, 1994 and 1995). This paper describes our recent efforts to extend Prophet along several dimensions: improved overhead control, greater resource sharing, greater resource heterogeneity, wide-area scheduling, and new application types. We show that both SPMD and task parallel applications can be scheduled effectively in a shared heterogeneous LAN environment containing ethernet and ATM networks by exploiting the application structure and dynamic run-time information.

Method and apparatus for allocating exclusive shared resource requests in a computer system

Abstract: A computer system incorporating an apparatus for allocating exclusive shared resource requests includes a shared resource, a first type device coupled to the shared resource and a second type device coupled to the shared resource. The computer system also includes an arbitrator unit coupled to the shared resource capable of granting the second type device exclusive access to the shared resource by preventing the first type device from being granted exclusive access to the shared resource. The arbitrator unit prevents the first type device from being granted exclusive access to the shared resource for at least a duration of time after the second type device has completed an associated second type shared resource transaction.

Network computer, and boot method applied to network computer

Abstract: A network computer, operating upon downloading programs and data necessary for data processing from a server computer, comprises a local storage device for storing programs and data downloaded from the server, and means for obtaining from the server an OS and information of setting up an environment for setting up working environment at the start of the network computer, and storing in the local storage device. By storing the OS and information of setting up the environment in the local storage device as described above, the network computer can be started from the local storage device at high speed without accessing the server, and a network computer suitable for mobile uses can be realized.

Enhancements to 4.4 BSD UNIX for efficient networked multimedia in project MARS

Abstract: Cluster based architectures that employ inexpensive personal computers (PCs) interconnected by high speed commodity interconnect have been recognized as a cost effective way of building high performance scalable multimedia-on-demand (MOD) storage servers (W. Bolosky et al., 1996; M. Buddhikot et al., 1994). Typically, the PCs in these architectures run operating systems such as UNIX that have traditionally been optimized for interactive computing and lack fast disk-to-network data paths and support for guaranteed CPU and storage access. We report design, implementation and performance measurements of innovative enhancements to 4.4 BSD UNIX carried out to rectify these limitations in the context of our Massively-parallel And Real-time Storage (MARS) project (M. Buddhikot et al., 1994). We have proposed and implemented the following enhancements to a 4.4 BSD compliant public domain NetBSD UNIX operating system: (1) a new kernel buffer management system called Multimedia M-buf (mmbuf) which shortens the data path from a storage device to network interface; (2) fair queueing within the SCSI driver for equitable resource sharing between real time and non real time streams; and (3) integration of these new OS services with a CPU scheduling mechanism called Real Time Upcall (R. Gopal, 1996) and a software disk striping driver called Concatenated Disk (CCD). Our experimental results demonstrate that these enhancements provide throughput improvements and QOS guarantees on the data path from the disk to network.

Apparatus and method for system resource object deallocation in a multi-threaded environment

Abstract: A shared resource registry is created and maintained by the system to control allocation and deallocation of shared system resources. The shared resource registry will contain information regarding the allocation and control of shared system resources. By accessing shared system resources through the shared resource registry, control of the shared resources is vested in the system instead of individual objects or processes. When the system needs access to a shared resource, it can notify the shared resource registry regarding the desired resource and schedule the termination of the thread or process which has acquired the desired resource. The shared resource registry can then release the desired resource and make it available for another process.

Managing a shared resource in a multi-processor system

Abstract: A data processing system provides a method and apparatus for managing a shared resource between multiple processors. The data processing system includes a first processor for producing, or supplying, available sections of the shared resource, and a second processor for consuming, or allocating, the available sections within the system. The first and second processor share a memory containing a circular queue, a resource queue, for referencing available shared resource sections. The resource queue is initialized by storing a pointer to an available shared resource section in each entry of the resource queue. The second processor consumes an available section of shared resource by removing a resource pointer from a resource queue entry and sending a message signal to a mailbox message subsystem of the first processor. The first processor produces an additional section of available shared resource by servicing the resource message signal and adding a pointer to the additional shared resource section in an empty resource queue entry.

Computer network system and portable computer

Abstract: A computer network system in which an IP address is assigned not only automatically at a destination of a notebook size PC (portable personal computer) 4 , but also information relating to shared resources on the network is obtained at the site from a server 1 through a LAN circuit 2 , so that the notebook size PC 4 can be used by connecting to the network even in the absence of the administrator. Henceforth, when the user moves within the same building as his own office, it seems that the occasion of using the portable personal computer by connecting to the network such as the Internet and intranet increases. In such a case, connection and disconnection of the network may be repeated frequently, and setting for such operation is facilitated.

Lock mechanism for shared resources in a data processing system

Abstract: A shared resource lock mechanism is provided which enables processors in a multi-processor system which each share common resources to obtain locks on those resources using a transactions which minimizes the amount of time system resources are unavailable, while also allowing system resources to be available for other processing tasks.

System for negotiating access to a shared resource by arbitration logic in a shared resource negotiator

Abstract: A resource negotiation technique and apparatus which streamlines arbitration for access to a shared resource by centralizing arbitration for groups of shared resources such as control registers into an access register An accessing agent first writes a request to an appropriate bit of a resource negotiation register (RNR), and then reads back a grant status If the request for access to the shared resource is not successful in the first attempt, the requesting processor may continuously read the grant status until it is successful Alternatively, the resource negotiation register may cause an interrupt in the requesting processor upon grant of access to the shared resource A logic level indicating that access is denied generally indicates that another processor in the multi-processor system is currently granted access to the corresponding shared resource Once the processor requesting permission for writing to the shared resource has been granted permission, that winning processor is expected to complete its access to the shared resource, and then indicate back to the shared resource negotiator that access is no longer required, eg, by clearing the corresponding request bit

Windows NT Clustering Service

Abstract: The Windows NT Clustering Service supports high-availability file servers, databases, and generic applications and services. A cluster is a collection of computer nodes-independent, self-contained computer systems-that work together to provide a more reliable and powerful system than a single node. In general, the goal of a cluster is to distribute a computing load over several systems, without users or system administrators being aware of the independent systems running the services. The Windows NT Clustering Service detects and restarts failed hardware or software components or migrates the failed component's functionality to another node if local restart is not possible. It also offers a much simpler user and programming interface. Microsoft Cluster Service for Windows NT has been shipping for about a year on Windows NT version 4.0. The upcoming Windows NT 5.0 release of Windows NT Clustering Service will improve ease of use through a wizard that guides the user through the creation of cluster resources.