Proceedings of 1993 5th IEEE Symposium on Parallel and Distributed Processing

1-4 Dec. 1993

Filter Results

Displaying Results 1 - 25 of 112
  • Optimal information dissemination in Star and Pancake networks

    Publication Year: 1993, Page(s):720 - 723
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (388 KB)

    The paper presents a new decomposition technique for hierarchical Cayley graphs. This technique yields a very easy implementation of the divide and conquer paradigm for weak order problems on very complex architectures as the Star graph or the Pancake. As applications, we introduce asymptotically optimal algorithms for broadcasting, gossiping, and semigroup operations for such interconnection netw... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Incremental garbage collection for causal relationship computation in distributed systems

    Publication Year: 1993, Page(s):650 - 655
    Cited by:  Patents (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (479 KB)

    Many distributed applications require the knowledge of the causality relation induced by the computation. Reconstructing this relation appears to be an interesting tool for such applications, but a vector of size S - where S is the number of processes - must be attached to each event to achieve this reconstruction. This induces a large overhead in secondary memory. After defining special events of... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • On the granularity of events when modeling program executions

    Publication Year: 1993, Page(s):422 - 429
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (763 KB)

    In the context of parallel and distributed programming, program execution models are required to obtain a better understanding of the behavior, properties, and characteristics of parallel and distributed applications. One commonly adopted approach consists of representing an execution as a set of events and relations that capture the ordering of the events. The main disadvantages of most existing ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An efficient algorithm for group communication

    Publication Year: 1993, Page(s):226 - 232
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (651 KB)

    We present an algorithm for reliable group communication that guarantees atomicity and total ordering in message delivery. The algorithm has been designed to operate with general omission failures on top of any datagram subnetwork. It allows one to operate within large groups of processes without loss of efficiency and is suitable for an easy implementation. Processes decide in at most 2fk + 2 pro... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Proceedings of 1993 5th IEEE Symposium on Parallel and Distributed Processing

    Publication Year: 1993
    Request permission for commercial reuse | PDF file iconPDF (52 KB)
    Freely Available from IEEE
  • Recursive Diagonal Torus: an interconnection network for massively parallel computers

    Publication Year: 1993, Page(s):591 - 594
    Cited by:  Papers (13)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (308 KB)

    Recursive Diagonal Torus (RDT), a class of interconnection network is proposed for massively parallel computers with up to 216 nodes. By adding remote links to the diagonal directions of the torus network recursively, the RDT can realize a smaller diameter (e.g., it is 11 for 216 nodes) with smaller number of links per node (i.e., 8 links per node) than that of the hypercube.... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A processor allocation strategy using cube coalescing in hypercube multicomputers

    Publication Year: 1993, Page(s):596 - 605
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (748 KB)

    We propose a new processor allocation strategy for hypercube multicomputers. The allocation strategy is called cube coalescing (CC) strategy since it uses a cube coalescing method to find an available subcube which is formed by the non-neighbored nodes. The main advantage of this strategy is that the nodes which do not contribute to form a requested subcube can be filtered out by coalescing, and t... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An efficient storage protocol for distributed object oriented databases

    Publication Year: 1993, Page(s):606 - 610
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (348 KB)

    In object oriented databases, objects are connected to each other by various kinds of relationships and form a general directed graph. The idea of leveled clustering is utilized to take advantage of the relationships and solve the problem of object allocation and storage in a distributed object oriented database. Both nonreplicative and replicative algorithms are developed to allocate objects to d... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Extended distributed genetic algorithm for channel routing

    Publication Year: 1993, Page(s):726 - 733
    Cited by:  Papers (5)  |  Patents (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (576 KB)

    In this paper, we propose a new parallel genetic algorithm (GA), called extended distributed genetic algorithm (EDGA) for channel routing problem. The EDGA combines the advantages of previous parallel GA models, viz., master/slave GA model and distributed GA model. In EDGA, the root processor executes the conventional genetic algorithm with global selection on total population and the remaining pr... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Performance effects of synchronization in parallel processors

    Publication Year: 1993, Page(s):611 - 616
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (428 KB)

    We investigate synchronization activities in application executing on distributed-memory MIMD architectures. Three applications are used to quantify the performance impact of synchronization as the number of processors is increased. We also investigate the performance improvement possible when synchronization is supported in hardware. The results show that significant performance improvement can b... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Execution behavior analysis and performance improvement in shared-memory architectures

    Publication Year: 1993, Page(s):23 - 26
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (328 KB)

    Communications latency forms a major obstacle to effective parallel processing. The bottlenecks of interprocessor communication can be caused by characteristics of a particular architecture or a particular application, and especially by the relationship between the two. We believe that efficient parallel processing requires serious attention to this intersection of architecture and application. In... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Scalable architectures with k-ary n-cube cluster-c organization

    Publication Year: 1993, Page(s):780 - 787
    Cited by:  Papers (5)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (624 KB)

    Recent advancements in VLSI and packaging technologies demonstrate attractiveness in building scalable parallel systems using clustered configurations while exploiting communication locality. Clustered architectures using buses or MINs as the inter-cluster interconnection do not satisfy both the above objectives. This paper proposes a new class of k-ary n-cube cluster-c scalable architectures by c... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An efficient mapping of multilayer perceptron with backpropagation ANNs on hypercubes

    Publication Year: 1993, Page(s):368 - 375
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (604 KB)

    This paper proposes a parallel structure, the mesh-of-appendixed-trees (MAT), for efficient implementation of artificial neural networks (ANNs). Algorithms to implement both the recall and the training phases of the multilayer perceptron and backpropagation ANN model are provided. A recursive procedure for embedding the MAT structure into the hypercube topology is used as the basis for an efficien... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A data-parallel approach to the implementation of weighted medians technique on parallel/super-computers

    Publication Year: 1993, Page(s):734 - 737
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (316 KB)

    Nonlinear filters have been used in many signal processing applications, for example, to obtain optimum signal extraction or detection in the presence of random noise. The weighted median filter (WMF), of which the standard median is a special case, is a novel nonlinear technique designed for 2D image processing. A major advantage of the WMF is its flexibility in design to deal with a wide variety... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Compiling distribution directives in a Fortran 90D compiler

    Publication Year: 1993, Page(s):617 - 620
    Cited by:  Patents (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (300 KB)

    Data partitioning and mapping is one of the most important steps of writing a parallel program, especially a data parallel one. Recently, Fortran D, and subsequently, High Performance Fortran (HPF) have been proposed to allow users to specify data distributions and alignments for the arrays in their programs. The paper presents the design of the data partitioning module of Fortran 90D compiler tha... View full abstract»

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

    Publication Year: 1993, Page(s):28 - 36
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (664 KB)

    Meerkat is a distributed memory multicomputer architecture that scales to hundreds of processors. Meerkat uses at two dimensional passive backplane to connect nodes composed of processors, memory, and I/O devices. The interconnect is conceptually simple, inexpensive to design and build, has low latency, and provides high bandwidth on long messages. However, it does not scale to thousands of proces... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • On partially dilated multistage interconnection networks with uniform traffic and nonuniform traffic spots

    Publication Year: 1993, Page(s):788 - 795
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (544 KB)

    The effects of partial dilation in interconnection networks is studied. Uniform traffic, bit-reverse permutations and mixed traffic models are considered in Delta-networks with conflict resolution by a packet drop mechanism. 1024*1024 Delta-networks with different switch sizes were used to examine the effects of partial dilation through simulations. An upper bound of the network throughput under b... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Data partitioning for networked parallel processing

    Publication Year: 1993, Page(s):376 - 379
    Cited by:  Papers (3)  |  Patents (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (268 KB)

    The workstation model of parallel processing presents specific challenges caused by the latency of the communications network and the workload imbalance that arises from the heterogeneity of the nodes. Data partitioning is critically important for parallel processing in this environment. We mathematically characterize the communication costs for four data decomposition schemes: scatter, contiguous... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Becoming a better host through origami: A mesh can be more than rows and columns

    Publication Year: 1993, Page(s):276 - 283
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (668 KB)

    Endowing a communication network with the ability to realize arbitrary communication patterns is an expensive proposition, both in hardware and in system software. One might instead ask whether a system can be built that performs well for a given application program. In this paper we look at the question of when a set of communication patterns is suitable for fast realization on a given network. I... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Matching dissimilar images: model and algorithm

    Publication Year: 1993, Page(s):738 - 741
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (352 KB)

    This paper addresses a particular problem in the area of matching dissimilar images, which can be formalized to image transform and functional minimization process. Based on the additive measuring model for sequential X-ray images with grey level deviation and the finite element technique, a class of grey level correction function and a robust matching criterion are developed, and an adaptive hier... View full abstract»

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

    Publication Year: 1993, Page(s):56 - 63
    Cited by:  Papers (2)  |  Patents (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (608 KB)

    We examine the wormhole routing problem in terms of the "congestion" c and "dilation" d for a set of packet paths. We show, with mild restrictions, that there is a simple randomized algorithm for routing any set of P packets in O(cdη + cLηlog P) time, where L is the number of flits in a packet, and η = min {d,L]; only a constant number of flits are stored in each queue at any time. Usi... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Parallel execution of multiple sequential instruction streams

    Publication Year: 1993, Page(s):621 - 624
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (264 KB)

    Most of the existing loop partitioning schemes require the dependence distances between the iterations of the loops to be constants which means that these schemes will not be applicable when the dependence distance is variable. When each iteration of a loop is treated as a sequential instruction stream, parallel execution of these loops is possible if the data dependency between each iteration can... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Correctness of a directory-based cache coherence protocol: Early experience

    Publication Year: 1993, Page(s):37 - 44
    Cited by:  Papers (2)  |  Patents (3)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (600 KB)

    Cache coherence protocols of increasing complexities call for automated verification tools which are both efficient and reliable. Most current approaches can only verify protocols at a high level of abstraction and the model size is limited to a small number of interacting processes. By using a simple full-map directory scheme as example, we present a verification technique which is extremely effi... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The design and implementation of late binding in a distributed programming language

    Publication Year: 1993, Page(s):438 - 445
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (712 KB)

    In distributed application domains where data change rapidly, it is often desirable for programs to obtain the latest available data values to achieve accurate computations. Example applications are financial services and network management. Such data are logically shared by a network of programs. Unlike data in traditional databases, rapidly changing data are usually not lockable by (client) prog... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An adaptive system-level diagnosis approach for hypercube multiprocessors

    Publication Year: 1993, Page(s):460 - 467
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (632 KB)

    This paper proposes a hierarchical adaptive system-level diagnosis approach for hypercube systems. Three measures for diagnosis cost (diagnosis time, number of tests and number of test links) are analyzed for the proposed algorithm. It is proved that the diagnosis cost required by this algorithm is lower than in the previous diagnosis algorithms in most of the fault cases. It is shown that the dia... View full abstract»

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