By Topic

Distributed Computing Systems, 1993., Proceedings the 13th International Conference on

Date 25-28 May 1993

Filter Results

Displaying Results 1 - 25 of 64
  • Proceedings of IEEE International Conference on Distributed Computing Systems - ICDCS '93

    Publication Year: 1993
    Save to Project icon | Request Permissions | PDF file iconPDF (27 KB)  
    Freely Available from IEEE
  • Practical considerations for non-blocking concurrent objects

    Publication Year: 1993 , Page(s): 264 - 273
    Cited by:  Papers (5)  |  Patents (39)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (848 KB)  

    An important class of concurrent objects are those that are nonblocking, that is, whose operations are not contained within mutually exclusive critical sections. A nonblocking object can be accessed by many threads at a time, yet update protocols based on atomic compare-and-swap operations can be used to guarantee the object's consistency. The author examines the compare-and-swap operation in the content of contemporary bus-based shared memory multiprocessors, although the results generalize to distributed shared memory multiprocessors. He describes an operating system-based solution that permits the construction of a nonblocking compare-and-swap function on architectures that only support more primitive atomic primitives such as test-and-set or atomic exchange. Several locking strategies are evaluated that can be used to synthesize a compare-and-swap operation, and it is shown that the common techniques for reducing synchronization overhead in the presence of contention are inappropriate when used as the basis for nonblocking synchronization. A simple synchronization strategy is described that has good performance because it avoids much of the synchronization overhead that normally occurs when there is contention View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Composition of concurrent programs

    Publication Year: 1993 , Page(s): 391 - 398
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (568 KB)  

    A model and a notation are developed for specifying the composition of concurrent programs. The work is based on the observation that the composition of concurrent programs often requires not only intraprocessor coordination but also interprocessor coordination. A notation is developed for explicitly specifying both forms of coordination within a single uniform framework. Much prior work has either ignored the interprocessor coordination aspects of composition, or treated it in a manner separate from the intraprocessor coordination aspects View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Decentralized consensus protocols with multi-port communication

    Publication Year: 1993 , Page(s): 356 - 365
    Cited by:  Papers (3)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (680 KB)  

    The authors develop efficient decentralized consensus protocols for a distributed system with multi-port communication. Two classes of decentralized consensus protocols are considered: the one without an initiator and the one with an initiator. The case of one-port communication is first presented, i.e., each node can send out one message in one step, and then results are derived for the case of multi-port communication, i.e., each node can send out more than one message in one step. Given an arbitrary number of nodes in a system, the proposed protocols can reach the consensus in the minimal numbers of message steps. The number of messages incurred by each algorithm is also derived View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Distribution and inheritance in the HERON approach to heterogeneous computing

    Publication Year: 1993 , Page(s): 399 - 408
    Cited by:  Papers (2)  |  Patents (8)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (788 KB)  

    HERON is a platform for object-oriented distributed computing in an open systems environment. We try to achieve a degree of distribution transparency previously known only from special distributed programming systems, while at the same time accommodating heterogeneous, autonomous computer systems. Distributed programs are written in Eiffel. The Eiffel language system is not modified: HERON employs proxies for remote object invocation and a flexible configuration procedure for building servers and distributed programs. In addition to regular objects, two kinds of distributed objects are supported by the proxy generator: dispersed objects and objects fragmented by remote inheritance. They contribute to distribution transparency both for distributed programs and for client/server systems View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • IPL: a multidatabase transaction specification language

    Publication Year: 1993 , Page(s): 439 - 448
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (900 KB)  

    A multidatabase system (MDBS) integrates preexisting and heterogeneous databases in a distributed environment. A multidatabase transaction is a consistent and reliable execution of an application over a multidatabase system. The authors summarize the characteristics of multidatabase transactions and present a multidatabase transaction specification language, the InterBase Parallel Language (IPL). IPL allows users to write MDBS transactions by specifying all associated actions, their sequences, control flow, and data flow among subtransactions, and yet retaining the autonomies of the preexisting software systems. IPL also allows users to specify different commit protocols for different subtransactions and to control the atomicity and isolation granularity of an MDBS transaction. IPL components and design issues are described in detail. The implementation of IPL is also discussed View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Using group communication to implement a fault-tolerant directory service

    Publication Year: 1993 , Page(s): 130 - 139
    Cited by:  Papers (5)  |  Patents (10)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (984 KB)  

    Group communication is an important paradigm for building distributed applications. The authors discuss a fault-tolerant distributed directory service based on group communication, and compare it with the previous design and implementation based on remote procedure call (RPC). The group directory service uses an active replication scheme and, when triplicated, can handle 627 lookup operations per second and 88 update operations per second (using nonvolatile RAM). This performance is better than the performance for the RPC implementation and it is even better than the performance for directory operations under SunOS, which does not provide any fault tolerance at all. The conclusion is that the implementation using group communication is simpler and has better performance than the one based on remote procedure call, supporting the claim that a distributed operating system should provide both remote procedure call and group communication View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Degradable agreement in the presence of Byzantine faults

    Publication Year: 1993 , Page(s): 237 - 244
    Cited by:  Papers (1)  |  Patents (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (640 KB)  

    The authors consider a system consisting of a sender that wants to send a value to certain receivers. Byzantine agreement protocols have previously been proposed to achieve this in the presence of arbitrary failures. The imposed requirement typically is that the fault-free receivers must all agree on the same value. An agreement protocol is proposed that achieves Lamport's Byzantine agreement (L. Lamport et al., 1982) up to a certain number of faults and a degraded form of agreement with a higher number of faults. The degraded form of agreement allows the fault-free receivers to agree on at most two different values, one of which is necessarily the default value. The proposed approach is named degradable agreement. An algorithm for degradable agreement is presented along with bounds on the number of nodes and network connectivity necessary to achieve degradable agreement View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Approximate analysis of priority scheduling systems using stochastic reward nets

    Publication Year: 1993 , Page(s): 466 - 473
    Cited by:  Papers (4)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (660 KB)  

    Presents a performance analysis of a heterogeneous multiprocessor system where tasks may arrive from Poisson sources as well as by spawning and probabilistic branching of other tasks. Non-preemptive priority scheduling is used between different tasks. Stochastic reward nets are used as the system model, and are solved analytically by generating the underlying continuous-time Markov chain. An approximation technique is used, that is based on fixed-point iteration to avoid the problem of a large underlying Markov chain. The iteration scheme works reasonably well, and the existence of a fixed point for the iterative scheme is guaranteed under certain conditions View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Deadline assignment in a distributed soft real-time system

    Publication Year: 1993 , Page(s): 428 - 437
    Cited by:  Papers (24)  |  Patents (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (788 KB)  

    In a distributed environment, tasks often have processing demands on multiple sites. A distributed task is usually divided up into several subtasks, each one to be executed at some site in order. In a real-time system, an overall deadline is usually specified by an application designer indicating when a distributed task is to be finished. To study the subtask deadline assignment problem a simple model of the system and tasks is postulated. The focus is on soft real-time systems. In such systems, it is very difficult to guarantee that all deadlines will be met, and hence one tries to minimize the number of deadlines that are missed. The authors examine (through simulations) four strategies for subtask deadline assignment in a distributed soft real-time environment View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Distributed application framework for large scale distributed systems

    Publication Year: 1993 , Page(s): 31 - 38
    Cited by:  Patents (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (616 KB)  

    The author introduces an infrastructure to develop large scale distributed applications, called the distributed application framework (DAF). With DAF, services provided by a distributed system are implemented as individual shared libraries that could be linked and loaded into a running process. A generic server is able to dynamically link these libraries to provide those services to its clients. To manage services in a system, DAF applies the notion of name space used by distributed file systems. That is, each server provides a Unix-filesystem-like name space to name and manage available services. The result is that managing services on a server is similar to handling files on a file system. A prototype of DAF has been implemented on top of Sun OS 4.1 using the Sun remote procedure call (RPC), external data representation (XDR), and dynamic linker View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Intelligent job selection for distributed scheduling

    Publication Year: 1993 , Page(s): 517 - 524
    Cited by:  Papers (4)  |  Patents (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (668 KB)  

    A key issue in distributed scheduling is selecting appropriate jobs to transfer. A job selection policy that considers the diversity of job behaviors is proposed. A mechanism used in artificial neural networks, called weight climbing, is employed. Using this mechanism, a distributed scheduler can learn the behavior of a job from its past executions and make a correct prediction about whether transferring the job is worthwhile. A scheduler using the proposed job selection policy has been implemented and experimental results show that it is able to learn job behaviors fast, make decisions accurately and adjust itself promptly when system configuration or program behaviors are changed. In addition, the selection policy introduces only negligible time and space overhead View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • General structured voting: a flexible framework for modelling cooperations

    Publication Year: 1993 , Page(s): 227 - 236
    Cited by:  Papers (7)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (848 KB)  

    Data replication schemes and mutual exclusion protocols can be regarded as special instances of cooperation schemes, which describe the interaction of independent nodes within a computer network to achieve a common goal. The author presents a new mechanism called general structured voting for cooperation management and demonstrates its use for handling instances of these problem domains. The proposed approach is shown to be very flexible, covers a wide range of scenarios, and supports an easy tailoring. In particular, it supports the switching from one cooperation scheme to another, e.g., as a reaction to changing network characteristics, simply by modifying the parameters of the model, avoiding a time and money consuming modification of the implementation View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Development of a collaborative application in CSDL

    Publication Year: 1993 , Page(s): 210 - 217
    Cited by:  Papers (5)  |  Patents (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (680 KB)  

    Cooperative system programming deals with four topics: multiuser interfaces, coordination, shared workspace, and networking control. The goal of CSDL (Cooperative Systems Design Language) is to cover all these aspects. The authors present the development of a system in CSDL. The system allows a group of physically distributed users to edit a document concurrently. It permits sharing the single-user editor xedit by multiplexing the application's outputs to each participant, while inputs come from one user at a time. A simple floor control policy allows participants to designate who has that right. A detailed presentation of the coordination layer, and a discussion of system architecture are included View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Sharing complex objects in a distributed PEER environment

    Publication Year: 1993 , Page(s): 186 - 193
    Cited by:  Papers (1)  |  Patents (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (796 KB)  

    For distributed computing environments, required for computer integrated manufacturing and other engineering applications, it is most important to support the sharing and exchange of complex objects among cooperating sites, while preserving their autonomy. Specification of complex objects and their object boundaries in a federated database are described. Each database, as well as the entire federation, is modeled as a collection of related objects. Complex objects are defined as subgraphs of the entire object base and are specified by a root object and a collection of paths. A complex object can be distributed over several sites. A method is described that ensures referential integrity while maintaining the autonomy of each database. Different linearization techniques of complex objects are supported to enable applications to retrieve complex objects as single entities. This model is implemented in PEER, a federated, object-oriented database system developed for engineering applications View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A general architecture for load balancing in a distributed-memory environment

    Publication Year: 1993 , Page(s): 47 - 54
    Cited by:  Papers (9)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (736 KB)  

    The goal of load balancing is to assign to each node a number of tasks proportional to its performance. On distributed-memory machines, it is important to take data dependencies into account when distributing tasks, since they have a big impact on the communication requirements of the distributed application. The authors present a load balancing architecture that can deal with applications with heterogeneous tasks. The idea is to provide a set of load balancers that are effective for different types of homogeneous tasks, and to allow users to combine these load balancers for applications with heterogeneous tasks. This architecture was implemented on the Nectar multicomputer and performance results are presented for several applications with homogeneous and heterogeneous tasks View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A subsystem for swapping and mapped file I/O on top of Chorus

    Publication Year: 1993 , Page(s): 12 - 19
    Cited by:  Patents (5)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (616 KB)  

    Chorus is a micro-kernel-based distributed operating system architecture. The authors explore the architectural and implementational issues involved in constructing a distributed paging service in the Chorus environment. Apart from outlining the pager architecture, they provide insight into how the characteristic goals of a critical distributed application on top of the Chorus system may be put into practice. The respective Chorus features are thereby judged in view of their suitability with respect to the pager implementation. The results of an experimental evaluation of the pager are included View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Coherence in naming in distributed computing environments

    Publication Year: 1993 , Page(s): 83 - 92
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (844 KB)  

    Many different kinds of names (identifiers) are used in computer systems. Names are resolved (interpreted) in a context. A context is a function that maps names to entities. Multiple contexts allow the flexibility of giving different meanings to a name in different parts of the system; however, there are situations where it is desirable for the meaning of a name to be the same in different parts. This property is called coherence in naming. Since the meaning of a name depends on the context selected, the analysis of coherence is based on the notion of closure mechanisms-implicit rules that select a context for resolving names. The authors define coherence and show how it is affected by various closure mechanisms. Then they present several approaches for dealing with the lack of coherence. Incoherence arises from selecting an incorrect context, and consequently, closure mechanisms are involved in the solutions View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Providing performance guarantees in an FDDI network

    Publication Year: 1993 , Page(s): 328 - 336
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (744 KB)  

    A network subsystem supporting a continuous media file system must guarantee a minimum throughput, a maximum delay, and a maximum jitter. The authors present a transport protocol that provides these guarantees. To support different types of service, the protocol is built from modules selected to meet the requirements of each communication session. A buffering technique is used to provide jitter guarantees. To provide throughput and delay guarantees, network performance is optimized based on the required transfer rate. The effects of controlling transmission rate and packet size are presented. The resulting transport protocol is modeled on a simulated FDDI (fiber distributed data interface) network and the results are analyzed. It is shown that the protocol provides the required guarantees for the anticipated types of traffic View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Deadlock prevention in the RTC programming system for distributed real-time applications

    Publication Year: 1993 , Page(s): 420 - 427
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (656 KB)  

    The RTC distributed real-time programming system was implemented using AND-OR locking of system resources to meet real-time and concurrency control requirements. Since RTC processes can hold locks while acquiring others, deadlock is possible and therefore a deadlock prevention technique was implemented for AND-OR locking in such systems. The authors briefly discuss the RTC programming system, illustrate the system's use in programming a timed version of the classic dining philosophers example, describe the deadlock prevention technique, and show how it is applied in the RTC dining philosophers example View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Maintaining information about persistent replicated objects in a distributed system

    Publication Year: 1993 , Page(s): 491 - 498
    Cited by:  Papers (5)  |  Patents (39)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (648 KB)  

    Presents a general model for persistent replicated object management and identify what metainformation about objects needs to be maintained by a naming and binding service to ensure that objects named by application programs are bound to only those object replicas which are in a mutually consistent state. These ideas are developed within the framework of a distributed system in which application programs are composed of atomic actions (atomic transactions) manipulating persistent (long-lived) objects View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Reconfiguration of spanning trees in networks in the presence of node failures

    Publication Year: 1993 , Page(s): 219 - 226
    Cited by:  Papers (3)  |  Patents (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (680 KB)  

    Connectivity among a set of user entities in a network can be provided by a network level abstraction of an acyclic graph (or spanning tree). The authors discuss the reconfiguration of a graph in the presence of failures of network nodes. A reconfiguration manifests itself as a graph fragmentation problem, whereby two or more disjoint subgraphs attempt to connect with one another to form a composite graph. Fragment interconnection requires contention resolution between fragments to avoid cycles. Two classes of contention resolution algorithms applicable for environments with a potentially large number of fragments are presented. They are based on preestablished ranking among fragments and random arbitration among fragments. The algorithms have been evaluated by simulation and compared. The algorithms are useful in supporting data multicasting across workstations and distributed computations involving data on different machines View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Uniform reliable multicast in a virtually synchronous environment

    Publication Year: 1993 , Page(s): 561 - 568
    Cited by:  Papers (22)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (644 KB)  

    The authors present the definition of and solution to the uniform reliable multicast problem in the virtually synchronous environment defined by the Isis system. A uniform reliable multicast of a message m has the property that if m has been received by any destination process (faulty or not), then m is received by all processes that reach a decision. Uniform reliable multicast provides a solution to the distributed commit problem. Two multicast primitives are defined in the virtually synchronous model: reliable multicast (called view-atomic) and uniform reliable multicast (called uniform view-atomic). The view-atomic multicast is used to implement the uniform view-atomic primitive. As view-atomicity is based on the concept of process group membership, a connection is established between the process group membership and the distributed commit problems View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Distributed shared repository: a unified approach to distribution and persistency

    Publication Year: 1993 , Page(s): 20 - 29
    Cited by:  Papers (4)  |  Patents (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (824 KB)  

    The authors propose an information management system providing distribution and persistency. By separating context from virtual address space, the system has a unified approach for both distribution and persistency. The former is achieved by moving contents between sites and the latter by moving contents between virtual address space and persistent storage. Contents include any information including data, program, and even the state of execution of a program. Contents are stored persistently in a logical space termed the distributed shared repository (DSR). A programming model for the DSR is proposed. Using the model, persistency, fine-grain mobility of information, and the passing of various distributed parameters can be obtained. The implementation anti experimental performance of the system are also presented View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Termination detection in a very general distributed computing model

    Publication Year: 1993 , Page(s): 374 - 381
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (640 KB)  

    Termination detection constitutes one of the basic problems of distributed computing, and many distributed algorithms have been proposed to solve it, but all these algorithms consider a very simple model for the underlying application programs: for processes of such programs, nondeterministic constructs are allowed, but each `receive' statement (request) concerns only one message at a time. A more realistic and very general model of distributed computing is first presented, allowing a request to be atomic on several messages and to obey AND/OR/AND-OR/k-out-of-n/etc. request types. Within this framework, two definitions of termination are proposed and discussed. Then, accordingly, two distributed algorithms for detecting these terminations are presented and evaluated; they differ in the information they use and in the time they need to claim termination View full abstract»

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