By Topic

IEEE Transactions on Parallel and Distributed Systems

Issue 2 • Date Feb 1997

Filter Results

Displaying Results 1 - 11 of 11
  • Graceful degradation in algorithm-based fault tolerant multiprocessor systems

    Publication Year: 1997, Page(s):137 - 153
    Cited by:  Papers (8)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (540 KB)

    Algorithm-based fault tolerance (ABFT) is a technique which improves the reliability of a multiprocessor system by providing concurrent error detection and fault location capability to it. It encodes data at the system level and modifies the algorithm to operate on the encoded data in order to expose both transient and permanent faults in any processor. Work done till now in this area takes care o... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A general method for maximizing the error-detecting ability of distributed algorithms

    Publication Year: 1997, Page(s):164 - 172
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (336 KB)

    The bound on component failures and their spatial distribution govern the fault tolerance of any candidate error-detecting algorithm. For distributed memory multiprocessors, the specific algorithm and the topology of the processor interconnection network define these bounds. This paper introduces the maximal fault index, derived from the system topology and local communication patterns, to demonst... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Algorithmic mapping of feedforward neural networks onto multiple bus systems

    Publication Year: 1997, Page(s):130 - 136
    Cited by:  Papers (3)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (208 KB)

    This paper addresses the problem of mapping a feedforward ANN onto a multiple bus system, MBS, with p processors and b buses so as to minimize the total execution time. We present an algorithm which assigns the nodes of a given computational layer (c-layer) to processors such that the computation lower bound [Nl/p]tpl and the communication lower bound [N... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Optimal task assignment in homogeneous networks

    Publication Year: 1997, Page(s):119 - 129
    Cited by:  Papers (21)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (560 KB)

    This paper considers the problem of assigning the tasks of a distributed application to the processors of a distributed system such that the sum of execution and communication costs is minimized. Previous work has shown this problem to be tractable for a system of two processors or a linear array of N processors, and for distributed programs of serial parallel structures. Here we focus on the assi... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The cross product of interconnection networks

    Publication Year: 1997, Page(s):109 - 118
    Cited by:  Papers (24)  |  Patents (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (256 KB)

    We study the cross product as a method for generating and analyzing interconnection network topologies for multiprocessor systems. Consider two interconnection graphs G1 and G2 each with some established properties such as symmetry, low degree and diameter, scalability, simple optimal routing, recursive structure (partitionability), fault tolerance, existence of node-disjoint... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Fusion of loops for parallelism and locality

    Publication Year: 1997, Page(s):193 - 209
    Cited by:  Papers (25)  |  Patents (6)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (564 KB)

    Loop fusion improves data locality and reduces synchronization in data-parallel applications. However, loop fusion is not always legal. Even when legal, fusion may introduce loop-carried dependences which prevent parallelism. In addition, performance losses result from cache conflicts in fused loops. In this paper, we present new techniques to: (1) allow fusion of loop nests in the presence of fus... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An empirical evaluation of performance-memory trade-offs in time warp

    Publication Year: 1997, Page(s):210 - 224
    Cited by:  Papers (5)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (536 KB)

    The performance of the Time Warp mechanism is experimentally evaluated when only a limited amount of memory is available to the parallel computation. An implementation of the cancelback protocol is used for memory management on a shared memory architecture, viz., KSR to evaluate the performance vs. memory tradeoff. The implementation of the cancelback protocol supports canceling back more than one... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Reducing communication latency with path multiplexing in optically interconnected multiprocessor systems

    Publication Year: 1997, Page(s):97 - 108
    Cited by:  Papers (16)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (352 KB)

    Reducing communication latency, which is a performance bottleneck in optically interconnected multiprocessor systems, is of prominent importance. A conventional approach for establishing connections in multiplexed networks uses a set of independent time slots (or virtual channels) along a path for each connection. This approach requires the use of switching devices capable of interchanging time sl... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Uniform and self-stabilizing token rings allowing unfair daemon

    Publication Year: 1997, Page(s):154 - 163
    Cited by:  Papers (7)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (256 KB)

    A distributed system consists of a set of processes and a set of communication links, each connecting a pair of processes. A distributed system is said to be self-stabilizing if it converges to a correct system state no matter which system state it starts with. A self-stabilizing system is considered to be an ideal fault tolerant system, since it tolerates any kind and any finite number of transie... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Multiple-edge-fault tolerance with respect to hypercubes

    Publication Year: 1997, Page(s):187 - 192
    Cited by:  Papers (6)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (392 KB)

    Previous works on edge-fault tolerance with respect to hypercubes Qn are mainly focused on 1-edge fault and 2- or 3-edge fault with limited size of n. We give a construction scheme for 2-EFT(Qn ) graphs and 3-EFT(Qn) graphs, where n is arbitrarily large. In our constructions, approximately log n extra degree is added to the vertices of Qn for 2-edge-faul... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • On runtime parallel scheduling for processor load balancing

    Publication Year: 1997, Page(s):173 - 186
    Cited by:  Papers (18)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (980 KB)

    Parallel scheduling is a new approach for load balancing. In parallel scheduling, all processors cooperate to schedule work. Parallel scheduling is able to accurately balance the load by using global load information at compile-time or runtime. It provides high-quality load balancing. This paper presents an overview of the parallel scheduling technique. Scheduling algorithms for tree, hypercube, a... View full abstract»

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

Aims & Scope

IEEE Transactions on Parallel and Distributed Systems (TPDS) is published monthly. It publishes a range of papers, comments on previously published papers, and survey articles that deal with the parallel and distributed systems research areas of current importance to our readers. Particular areas of interest include, but are not limited to:

Full Aims & Scope

Meet Our Editors

Editor-in-Chief
David Bader
College of Computing
Georgia Institute of Technology