Proceedings of the Third IEEE Symposium on Parallel and Distributed Processing

2-5 Dec. 1991

Filter Results

Displaying Results 1 - 25 of 119
  • Data communication and computational geometry on the star and pancake interconnection networks

    Publication Year: 1991, Page(s):415 - 422
    Cited by:  Papers (19)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (698 KB)

    The star and pancake networks were recently proposed as attractive alternatives to the hypercube topology for interconnecting processors in a parallel computer. However, little has been done to design parallel algorithms on these networks. The paper presents several data communication algorithms that are fundamental to designing algorithms on these two networks. These algorithms are then used to d... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Proceedings of the Third IEEE Symposium on Parallel and Distributed Processing (Cat. No.91TH0396-2)

    Publication Year: 1991
    Request permission for commercial reuse | |PDF file iconPDF (35 KB)
    Freely Available from IEEE
  • A high speed systolic architecture for labeling connected components in an image

    Publication Year: 1991, Page(s):818 - 825
    Cited by:  Papers (3)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (540 KB)

    Connected components detection and labeling is an essential step in many image analysis techniques. The efficiency of the connected components labeling algorithm is critical for many image processing and machine vision applications that require real time response. The advances in the areas of parallel processing and VLSI technology can be exploited in designing hardware algorithms of high speed an... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Broadcasting on incomplete hypercubes

    Publication Year: 1991, Page(s):826 - 833
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (628 KB)

    Incomplete hypercubes make the hypercubes more flexible on task allocation in large cubes, cost of manufacturing hardware, and hypercubes with faulty nodes. The authors devise and analyze broadcasting algorithms based on the hierarchical binomial spanning trees, hierarchical edge-disjoint spanning trees and edge-disjoint spanning trees in an incomplete hypercube of 2n+2k node... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Packet routing on square meshes with row and column buses

    Publication Year: 1991, Page(s):834 - 837
    Cited by:  Papers (10)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (300 KB)

    General point-to-point communication among processors in the classical two-dimensional n×n square mesh architecture necessarily takes at least 2n-2 time steps. The authors investigate the problem of routing arbitrary permutations on an enhanced square mesh with separate broadcast buses along each of its rows and columns. They prove that any packet routing algorithm... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Parallel tree search on a SIMD machine

    Publication Year: 1991, Page(s):249 - 256
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (636 KB)

    Presents an implementation of a depth-first heuristic tree search on the single-instruction, multiple-data (SIMD) Connection Machine. The algorithm is based on Iterative-Deepening-A*. Until recently, only highly regular, data-parallel computations have been performed on SIMD machines. Searching an irregular tree represents a new application of SIMD machines. The main technical challenge is load ba... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Implementing a conventional language for a dataflow architecture

    Publication Year: 1991, Page(s):164 - 171
    Cited by:  Papers (4)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (684 KB)

    Details the implementation of the Pascal programming language for a dataflow architecture. The reasons for choosing this particular architectural model and language was to achieve maximum parallelism with minimal specification by the programmer. Issues that are discussed involve the analysis and transformation of code to maximise parallelism, and the generation of an applicative intermediate code ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Upper and lower bounds for one-write multivalued regular registers

    Publication Year: 1991, Page(s):134 - 141
    Cited by:  Papers (2)  |  Patents (1)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (680 KB)

    This paper presents an algorithm for implementing a k-ary regular register (the logical register) using k(k-1)/2 binary regular registers (the physical registers) that requires only one physical write per logical write. The algorithm is simple to describe and depends on properties of paths in a related graph. Two lower bounds on the number of registers required by one-wr... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Deadlock-free adaptive routing algorithms for multicomputers: evaluation of a new algorithm

    Publication Year: 1991, Page(s):840 - 847
    Cited by:  Papers (32)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (740 KB)

    A new theory for the design of deadlock-free adaptive routing algorithms for wormhole and store-and-forward routing as well as two design methodologies have been proposed previously. In this paper, a new adaptive routing algorithm, obtained from the application of the former theory to the binary n-cube, is evaluated under different load conditions and compared with other algorithms. The r... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Approximate MVA from Markov model of software client/server systems

    Publication Year: 1991, Page(s):322 - 329
    Cited by:  Papers (6)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (628 KB)

    Stochastic rendezvous networks (SRVN) are performance models for multitasking parallel software with intertask communication via rendezvous (RV). The paper describes an approximation for the arrival instant probabilities for a class of simple client/server SRVN, based on a rigorous analysis of the Markov chain model describing the interference of different client tasks contending for a single serv... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Parallel algorithm and complexity results for telephone link simulation

    Publication Year: 1991, Page(s):378 - 385
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (556 KB)

    The telephone connection problem (TCP) is the problem of simulating a telephone link of fixed capacity to assess its ability to serve incoming calls. This simulation is performed on a large number of sample calls at AT&T Bell Laboratories. In order to speed up the simulation, it is desirable to obtain good parallel algorithms for the problem. The authors give an O(k log n) time parallel on an ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An optimal algorithm for the longest common subsequence problem

    Publication Year: 1991, Page(s):630 - 639
    Cited by:  Papers (3)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (868 KB)

    The longest common subsequence problem is to find a longest common subsequence of two given strings. The complexity of this problem on the decision tree model is known as mn, where m and n are the lengths of these two strings, respectively, and m⩽n. The authors present a parallel algorithm for this problem on the CREW PRAM model, which takes O(log2mloglogm) time with mn/log2 View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A shared-memory multiprocessor garbage collector and its evaluation for committed-choice logic programs

    Publication Year: 1991, Page(s):870 - 877
    Cited by:  Patents (1)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (716 KB)

    A parallel copying garbage collection algorithm for symbolic languages executing on shared-memory multiprocessors is proposed. The algorithm is an extension of Baker's sequential algorithm with a novel method of heap allocation to prevent fragmentation and facilitate load distribution during garbage collection. An implementation of the algorithm within a concurrent logic programming system, VPIM, ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The M-cache: a message-retrieving mechanism for multicomputer systems

    Publication Year: 1991, Page(s):258 - 265
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (572 KB)

    Presents the design and evaluation of the M-cache, a small, fast and intelligent memory for handling messages at the processing nodes of multicomputer systems. The M-cache provides hardware support for the message search operation often performed in message-directed programming. It also provides a mechanism for bandwidth matching between the interconnection network and local memory of a node. Thro... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A comparison of structure accessing techniques in IdA and SISAL on the CSIRAC II dataflow multiprocessor

    Publication Year: 1991, Page(s):172 - 179
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (724 KB)

    Describes the implementation of the functional language IdA on the CSIRAC II dataflow multiprocessor. IdA is a derivative of MIT's ID Nouveau language and CSIRAC II is a dataflow architecture which combines the features of static queued and dynamic dataflow architectures and as such is outside the generally accepted taxonomy. The non-strict implementation of structures in IdA exploits the overlap ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Detection of strong predicates in distributed programs

    Publication Year: 1991, Page(s):692 - 699
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (572 KB)

    This paper discusses detection of unstable predicates in a distributed program. Some applications of this are in program debugging and testing. The authors provide a predicate logic in the form of a grammar giving the rewrite rules for constructing predicates about a distributed program. This predicate logic is general enough to describe many conditions programmers are interested in. These predica... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Explicit dynamic exclusion algorithm

    Publication Year: 1991, Page(s):142 - 149
    Cited by:  Papers (3)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (672 KB)

    An algorithm for the distributed (explicitly) dynamic resources allocation problem is presented. The algorithm is based on Chandy and Misra's Dining Philosophers Algorithm. Their algorithm is extended to allow philosophers to request that they be added or deleted from the problem. Philosophers are also allowed to add and delete resources from their set of needed resources. This solution can also b... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Embedding complete binary trees in faulty hypercubes

    Publication Year: 1991, Page(s):112 - 119
    Cited by:  Papers (3)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (636 KB)

    This paper studies the ability of the hypercube to implement tree-structured algorithms in the presence of faults. The hypercube is able to implement a wide range of algorithms efficiently, and the authors' selection of tree computations is motivated by the fact that many parallel algorithms, including broadcasting, parallel prefix, and other divide-and-conquer algorithms, have a natural tree stru... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An algorithm for dynamic reconfiguration of a multicomputer network

    Publication Year: 1991, Page(s):848 - 855
    Cited by:  Papers (4)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (680 KB)

    The dynamic reconfiguration of the interconnection network is an advanced feature of some multicomputers to reduce the communication overhead. The authors present an algorithm for the dynamic reconfiguration of the network. Reconfiguration is limited, preserving the original topology. Long distance message passing is minimized by positioning communication partners close to each other. The algorith... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Performance of multi-threaded execution in a shared-memory multiprocessor

    Publication Year: 1991, Page(s):330 - 333
    Cited by:  Papers (6)  |  Patents (12)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (340 KB)

    The authors propose a general analytic model for the overlapped execution of thread management operations in shared-memory multiprocessors. Performance measures as speedup, utilization and network contention are evaluated and verified using simulations. It is assumed that the execution of each thread consists of sequence of single-cycle instructions, interleaved with four high-level thread managem... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Lower bounds and parallel algorithms for planar orthogonal grid drawings

    Publication Year: 1991, Page(s):386 - 393
    Cited by:  Papers (3)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (508 KB)

    The paper considers the problem of constructing a planer orthogonal grid drawing (or more simply, layout) of an n-vertex graph, with the goal of minimizing the number of bends along the edges. It exhibits graphs that require Ω(n) bends in any layout, and shows that there exist optimal drawings that require Ω(n) bends and have all of them on a single edge of length Ω(n2... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An optimizing compiler for FP*-a data-parallel dialect of FP

    Publication Year: 1991, Page(s):70 - 78
    Cited by:  Patents (1)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (660 KB)

    The authors present an overview of a compiler for the functional language FP* that generates programs suitable for execution on massively parallel architectures. Data-parallel programs are naturally expressed using FP* functionals, which map efficiently onto massively parallel architectures. However, the creation of large and unnecessary intermediate results is a general source of inefficiencies i... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • On mapping Bezier curve and surface subdivision algorithms into a hypercube with geometric adjacency constraint

    Publication Year: 1991, Page(s):582 - 589
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (600 KB)

    The authors discuss the Bezier curve and surface generation algorithms on a hypercube computer. They show that the computation structures of Bezier curve and surface generation based on subdivision method can be modeled as binomial trees and extended binomial trees respectively. Properties of binomial trees and extended binomial trees are explored and mappings from these tree structures to hypercu... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • On a fast deterministic parallel approximate matching algorithm

    Publication Year: 1991, Page(s):774 - 777
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (288 KB)

    Sublinear parallel graph matching algorithms are investigated. The author proposes an approximation scheme for the cardinality matching problem on general graphs that runs in O(k5log4n) parallel time using O(2kn2k+2) processors on a CREW PRAM parallel machine model. The approximation is at least as big as View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Explicit construction for reliable reconfigurable array architectures

    Publication Year: 1991, Page(s):640 - 647
    Cited by:  Papers (3)
    Request permission for commercial reuse | Click to expandAbstract |PDF file iconPDF (604 KB)

    This paper describes some explicit constructions for reconfigurable array architectures. Given a working architecture (application graph), the authors add redundant hardware to increase reliability. The degree of reconfigurability, DR, of a redundant graph is a measure of the cost of reconfiguration after failures. When DR is independent of the size of the application graph, the authors say the gr... View full abstract»

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