By Topic

Distributed Computing Systems, 1995., Proceedings of the 15th International Conference on

Date May 30 1995-June 2 1995

Filter Results

Displaying Results 1 - 25 of 67
  • Proceedings of 15th International Conference on Distributed Computing Systems

    Save to Project icon | Request Permissions | PDF file iconPDF (378 KB)  
    Freely Available from IEEE
  • Future distributed embedded and real-time applications will be adaptive - meanings, challenges and research paradigms [panel session]

    Save to Project icon | Request Permissions | PDF file iconPDF (80 KB)  
    Freely Available from IEEE
  • Author index

    Save to Project icon | Request Permissions | PDF file iconPDF (97 KB)  
    Freely Available from IEEE
  • Write caching in distributed file systems

    Page(s): 457 - 466
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (868 KB)  

    Disk caches are employed in distributed file systems to avoid network accesses at clients and to compensate for the speed differential between main memory and disk at file servers. Because of concerns about volatility, however, write requests have typically not benefitted from the presence of caches. Instead, they have been processed with some sort of write-through or periodic write-back approach to ensure the integrity of the stored data. The introduction of reasonably priced non-volatile (NV) memories has prompted interest in the use of such memory for write caching, at the server and/or at the client. This paper describes an investigation through trace-driven simulation experiments of several approaches to write caching in distributed systems, with both volatile and non-volatile caches. The results support the findings of earlier work that suggests important differences between caching in the traditional single-level caching environment and caching in a two-level caching environment. While policies focusing on temporal locality perform well for a single-level caching system, or at the client of a two-level caching system, they may not be suitable for use at the server in a two-level caching system. This is because locality characteristics in the reference stream seen at the server in a two-level caching system may be destroyed by caching at the client with a NV write cache large enough to hold the client's working set of dirty blocks. Policies focusing on amortizing the cost of a disk seek operation over multiple write-back operations perform better at the server of a two-level caching system View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Parallel processing on networks of workstations: a fault-tolerant, high performance approach

    Page(s): 467 - 474
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (796 KB)  

    One of the most sought after software innovation of this decade is the construction of systems using off-the-shelf-workstations that actually deliver and even surpass, the power and reliability of supercomputers. Using completely novel techniques: eager scheduling, evasive memory layouts and dispersed data management it is possible to build an execution environment for parallel programs on workstation networks. These techniques were originally developed in a theoretical framework for an abstract machine which models a shared memory asynchronous multiprocessor. The network of workstations platform presents an inherently asynchronous environment for the execution of our parallel program. This gives rise to substantial problems of correctness of the computation and of proper automatic load balancing of the work amongst the processors, so that a slow processor will not hold up the total computation. A limiting case of asynchrony is when a processor becomes infinitely slow, i.e. fails. Our methodology copes with all these problems, as well as with memory failures. An interesting feature of this system is that it is neither a fault-tolerant system extended for parallel processing nor is it parallel processing system extended for fault tolerance. The same novel mechanisms ensure both properties View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Schedulability-oriented replication of periodic tasks in distributed real-time systems

    Page(s): 196 - 203
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (732 KB)  

    We consider the schedulability-oriented replication problem of a set of periodic real-time tasks where each task can be decomposed into several modules and intermodule communications. The objective is to find an allocation in which there exists a feasible schedule for the given task set. In this paper, we adopt a communication model where the replication of modules is not for the sake of fault tolerance but for increasing the degree of schedulability. To solve the problem, we develop a replication technique and embed the technique in a simulated annealing algorithm. Experimental results show that such replication may lead to a higher degree of schedulability and obtain a feasible solution View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Distributed management by delegation

    Page(s): 333 - 340
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (900 KB)  

    This paper introduces a novel approach to distributed computing based on delegation-agents, and describes its applications to decentralize network management. Delegation agents are programs that can be dispatched to remote processes, dynamically linked and executed under local or remote control. Unlike scripted agents, delegation agent programs may be written in arbitrary languages, interpreted or compiled. They can thus be more broadly applied to handle such tasks as real-time monitoring, analysis and control of network resources. Distributed management by delegation (MbD) uses this to manage remote elements and domains. MbD provides a paradigm for distributed, flexible, scalable and robust network management that overcomes the key limitations of current centralized management schemes View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Specification of a secured multi-server MMS protocol

    Page(s): 526 - 533
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (584 KB)  

    We present a complete specification of a new architecture for the Manufacturing Message Specification (MMS) protocol. This architecture is based on a client/multi-server model adopted by the MMS standard. A client invokes a service request that implies the cooperation of many servers to execute the requested service and satisfies the client demand. This architecture permits the definition of distributed objects, each of them is composed of simple MMS objects distributed over many servers. The coordination of the MMS services execution across the distributed cooperated servers as well as the coherence of the distributed object are guaranteed by the services and the protocol offered by the ISO Commitment, Concurrency, and Recovery (CCR) application service element. The specification of the secured MMS multi-server protocol is introduced by a finite state machine model. This model is used as a base for a formal validation using the Calculus of Communicating Systems (CCS) View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A distributed table-driven route selection scheme for establishing real-time video channels

    Page(s): 52 - 59
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (748 KB)  

    To guarantee the delivery of real-time messages before their deadline, a real-time connection or channel must be established before the transmission of any real-time messages. During this channel-establishment phase, one must first select a route between the source and destination of this channel and then reserve sufficient resources along this route so that the worst-case end-to-end delay over the selected route may not exceed the user-specified delay bound. We propose a table-driven distributed route-selection scheme that is guaranteed to find a “qualified” route, if any, that meets the performance requirement of the requested channel without compromising any of the existing guarantees. The proposed scheme uses the Bellman-Ford shortest path algorithm to build real-time delay tables, and hence, can solve the route-selection problem by a simple table look-up. Several examples are presented to demonstrate the effectiveness of the proposed distributed route-selection scheme View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Software tool evaluation methodology

    Page(s): 3 - 10
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (756 KB)  

    The recent development of parallel and distributed computing software has introduced a variety of software tools that support several programming paradigms and languages. This variety of tools makes the selection of the best tool to run a given class of applications on a parallel or distributed system a non-trivial task that requires some investigation. We expect tool evaluation to receive more attention as the deployment and usage of distributed systems increases. In this paper, we present a multi-level evaluation methodology for parallel/distributed tools in which tools are evaluated from different perspectives. We apply our evaluation methodology to three message passing tools viz Express, p4, and PVM. The approach covers several important distributed systems platforms consisting of different computers (e.g., IBM-SP1, Alpha cluster, SUN workstations) interconnected by different types of networks (e.g., Ethernet, FDDI, ATM) View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A distributed K-mutual exclusion algorithm

    Page(s): 153 - 160
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (636 KB)  

    This paper presents a token-based K-mutual exclusion algorithm. The algorithm uses K tokens and a dynamic forest structure for each token. This structure is used to forward token requests. The algorithm is expected to minimize the number of messages and also the delay in entering the critical section, at low as well as high loads. The paper presents simulation results for the proposed algorithm and compares them with three other algorithms. Unlike previous work, our simulation model assumes that a finite (non-zero) overhead is encountered when a message is sent or received. The simulation results show that, as compared to other algorithms, the proposed algorithm achieves lower delay in entering critical section as well as lower number of messages, without a significant increase in the size of the messages View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Causal separators for large-scale multicast communication

    Page(s): 83 - 91
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (836 KB)  

    In recent years there has been a growing interest in developing communication systems that are able to deliver messages respecting potential causality. Unfortunately, causal delivery cannot be provided without costs: extra delays may be induced on message delivery or processes may be required to maintain and exchange records of causal relations. In this paper we present an extension to previous work on compression of causal information using knowledge about the topology of the communication structure. In order to make practical use of this result, we present a methodology to model the communication system. The technique exploits the physical structure of existing networks, in particular its hierarchical nature, to create a communication graph where causal separators match the underlying physical and administrative organization. We show that this approach can be applied to existing large-scale systems, providing the means for using topological timestamping with negligible overhead View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Integrating visualization support into distributed computing systems

    Page(s): 19 - 26
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (800 KB)  

    Visualization and animation tools may become extremely important aids in the understanding, verification, and performance tuning of parallel computations. Presently, however, the use of visualization has had only a limited use for enhancing parallel computation. We hypothesize that one of the primary reasons for the limited use of visualization tools in parallel program development is the difficulty of acquiring the information necessary to drive the visual display. Our approach to this impediment focuses on integrating visualization support directly into a distributed computing system. Central to this integration is the addition of a logical clock that prevents the timestamps of events from violating causality. The implementation requires the “piggybacking” of a negligible amount of extra header information on system messages and the impact on performance is minimal. This results in a system that produces useful visualizations with no extra effort required by the applications programmer. Also integrated into the distributed system is support which simplifies the creation of programmer-defined, application-specific visualizations, unique to each new parallel program developed View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • “How are we going to pay for this? Fee-for-service in distributed systems-research and policy issues”

    Page(s): 344 - 348
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (268 KB)  

    With the increasing array of information and services being supported by distributed computing, we face a new challenge: How do we handle charges? Providers of information will want to receive “royalties”, providers of computing services will want to receive payment for use of the service, and providers of the network will want to receive payment for the transmission. At some point, the users of the provided services will have to pay for this. What do we need to do in designing our distributed computing systems to support some form of charge/payment? This paper discusses these issues View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Adaptive placement of method executions within a customizable distributed object-based runtime system: design, implementation and performance

    Page(s): 279 - 286
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (832 KB)  

    This paper presents the design and implementation of a mechanism aimed at enhancing the performance of distributed object-based applications. This goal as achieved by means of a new algorithm implementing placement of method executions that adapts to processors' load and to objects' characteristics, the latter allowing to approximate the cost of methods' remote execution. The behavior of the proposed placement algorithm is examined by providing performance measures obtained from its integration within a customizable distributed object-based runtime system. In particular, the cost of method executions using our algorithm is compared with the cost resulting from the standard placement technique that consists of executing any method on the storing node of its embedding object View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Fault-tolerant external clock synchronization

    Page(s): 70 - 77
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (676 KB)  

    We address the problem of how to integrate fault-tolerant internal and external clock synchronization. We propose a new algorithm which provides both external and internal clock synchronization for as long as no more than F reference time servers out of a total of 2F+1 are faulty. When the number of faulty reference time servers exceeds F, the algorithm degrades to a fault-tolerant internal clock synchronization algorithm. We prove that at least 2F+1 reference time servers are necessary for achieving external clock synchronization when up to F reference time servers can suffer arbitrary failures, thus our algorithm provides maximum fault-tolerance. The algorithm is also optimal in another sense: we show that the maximum deviation between reference time and the clocks of nonreference time servers is minimal View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A fast distributed modular algorithm for resource allocation

    Page(s): 161 - 168
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (652 KB)  

    This paper concerns resource allocation in distributed message passing systems, i.e., the scheduling of accesses to system resources shared among many concurrent processes. Three different kinds of resource allocation problems with varying degrees of generality are considered: the dining philosophers problem, the drinking philosophers problem and the dynamic resource allocation problem. We present an efficient modular resource allocation algorithm that uses any arbitrary resource allocation algorithm as a subroutine. It improves the performance of the subroutine by letting each process wait only for its currently conflicting processes, and therefore, allows more concurrency. For appropriate choices of the subroutine, we obtain the fastest known resource allocation algorithms in terms of the worst case response time. Simulation studies were conducted which also indicate that our algorithms perform faster and require a smaller number of messages than other previously known algorithms on average, especially when resource contention among processes is high and the average time that a process remains in the critical region is large View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Newtop: a fault-tolerant group communication protocol

    Page(s): 296 - 306
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1100 KB)  

    A general purpose group communication protocol suite called Newtop is described. It is assumed that processes can simultaneously belong to many groups, group size could be large, and processes could be communicating over the Internet. Asynchronous communication environment is therefore assumed where message transmission times cannot be accurately estimated, and the underlying network may well get partitioned, preventing functioning processes from communicating with each other. Newtop can provide causality preserving total order delivery to members of a group, ensuring that total order delivery is preserved for multi-group processes. Both symmetric and asymmetric order protocols are supported, permitting a process to use say symmetric version in one group and asymmetric version in other View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Coterie templates: a new quorum construction method

    Page(s): 92 - 99
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (568 KB)  

    One approach to distributed mutual exclusion algorithms is the use of quorums. Quorum-based algorithms offer the advantage of protocol symmetry, spreading effort and responsibility uniformly across the distributed system. In this paper, we present an O(logn) algorithm to generate coterie templates of near-optimal O(n0.63) size. Coterie templates are generic quorum structures that exhibit several desirable properties such as fault tolerance, symmetry and low storage cost. In addition, coteries can be instantiated from the template to reflect desirable network characteristics View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Configuration-level optimization of RPC-based distributed programs

    Page(s): 307 - 315
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (840 KB)  

    Many strategies for improving performance of distributed programs can be described abstractly in terms of an application's overall configuration. But previously those techniques would need to be implemented manually, and the resulting programs, though yielding good performance, are more expensive to build and much less easy to reuse. This paper describes research towards an automatic system for introducing performance improvement techniques based upon an application's configuration description View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Performance evaluation of three logging schemes for a shared-nothing database server

    Page(s): 221 - 228
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (812 KB)  

    The European Declarative System (EDS) is a high performance backend database server designed for a range of commercial mainframes. One major application domain of EDS is information processing for business and commercial environments. High performance is achieved by exploiting parallelism using a shared-nothing computer (up to 256 processors). Reliability is a crucial design issue for commercial and business information systems. Recovery control facilitates reliability and logging forms an important part of it. In general, logging is costly to implement. It is usually achieved at the expense of reduced system performance. Three logging schemes have been studied for EDS: (a) local discs-adopt a conventional approach by incorporating a local disc on each processor; (b) duplexing-arrange the processors in pairs, one for database operations and one for backup; and (c) cooperative logging-similar to duplexing except database and backup operations are performed on a single processor. The performance of these schemes for on-line transaction processing was evaluated and compared using the EDS behavioural simulator. The results of the evaluations are presented in this paper View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A practical technique for asynchronous transaction processing

    Page(s): 110 - 117
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (940 KB)  

    Asynchronous transaction processing extends traditional on-line transaction processing (TP) to improve performance of distributed systems by alleviating the serializability (SR) bottleneck. For example, epsilon serializability (ESR) uses divergence control algorithms to allow more concurrency by permitting limited non-SR interleavings. In a distributed environment, ESR relaxes commit and abort dependencies among transactions, allowing transactions to commit asynchronously. A second example, chopping up transactions allows more concurrency by dividing transactions into smaller pieces and thus reduces resource holding time. Chopping transactions enforces no commit protocols among pieces from one original transaction, allowing each piece to commit asynchronously. We combine the benefits of ESR and chopping transactions by designing three new methods that chop transactions and run them under ESR. The practical applicability of our technique is enhanced by two factors: (1) chopping transactions does not require changes in existing TP systems, and (2) ESR support has already been prototyped on a commercial TP system View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Single connection emulation (SCE): an architecture for providing a reliable multicast transport service

    Page(s): 144 - 151
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (744 KB)  

    We present a novel architecture for providing a reliable multicast transport service over existing protocol stacks. These protocol stacks ordinarily support reliable unicast transport layer connections over a network layer which is capable of providing an unreliable multicasting service. We propose the addition of a new single connection emulation (SCE) sublayer between the unicast transport layer and the multicast network layer. This added layer mimics the single destination network layer interface to the transport layer and interfaces with the multicast network layer to provide the necessary multicast functionality. The new architecture also enables interactions between applications and the SCE, thus allowing the applications to control the semantics of the reliable multicast connection. We discuss the design issues that need to be considered when such a sublayer is to be introduced. We also discuss an implementation of this new approach using the TCP/IP protocol stack and present some preliminary experimental results View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Distributed pipeline scheduling: end-to-end analysis of heterogeneous, multi-resource real-time systems

    Page(s): 204 - 211
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (668 KB)  

    This paper presents an hierarchical end-to-end analysis technique that decomposes the very complex heterogeneous multi-resource scheduling problem into a set of single resource scheduling problems with well defined interactions. We define heterogeneity both in resource types, e.g., CPU, and in scheduling policies, e.g., rate-monotonic scheduling. This analysis technique is one phase of our systems integration framework for designing large-scale, heterogeneous, distributed real-time systems whose timing properties can be strictly controlled and analyzed. This approach, denoted the Distributed Pipelining Framework, exploits the natural pipelining execution pattern found in a large number of continuous (periodic) applications executing over heterogenous resources. A teleconference application is used in this paper to show the utility of the approach View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • MASSIVE: a distributed virtual reality system incorporating spatial trading

    Page(s): 27 - 34
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (828 KB)  

    MASSIVE is a distributed virtual reality system. It provides rich facilities to support user interaction and cooperation via text, audio and graphics media and interaction is controlled by a spatial model of interaction. The communications architecture is based on processes communicating via typed connections which have interfaces on both ends and which integrate RPCs, attributes and streams in a common context. A spatial interface trading service, the aura manager, has been developed to support interface trading in a spatial context. The concepts embodied in the aura manager can be useful in other interface trading situations, especially where notions of “space” and “meeting” may be applied View full abstract»

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