By Topic

Parallel and Distributed Systems, IEEE Transactions on

Issue 12 • Date Dec. 1997

Filter Results

Displaying Results 1 - 14 of 14
  • Comments on "A new family of Cayley graph interconnection networks of constant degree four"

    Page(s): 1299 - 1300
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (77 KB)  

    For original paper see Vadapalli and Srimani, ibid., vol. 7, no. 1, p 26-32, 1996, where the authors have proposed a new family of Cayley graph interconnection networks of constant degree four. Our comments show that their proposed graph is not new but is the same as the wrap-around butterfly graph. The structural kinship of the proposed graph with the de Bruijn graph is also discussed. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • 1997 Index IEEE Transactions on Parallel And Vol. 8

    Page(s): 1301 - 1312
    Save to Project icon | Request Permissions | PDF file iconPDF (457 KB)  
    Freely Available from IEEE
  • Solving an algebraic path problem and some related graph problems on a hyper-bus broadcast network

    Page(s): 1226 - 1235
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (448 KB)  

    The parallel computation model upon which the proposed algorithms are based is the hyper-bus broadcast network. The hyper-bus broadcast network consists of processors which are connected by global buses only. Based on such an improved architecture, we first design two O(1) time basic operations for finding the maximum and minimum of N numbers each of size O(log N)-bit and computing the matrix multiplication operation of two N×N matrices, respectively. Then, based on these two basic operations, three of the most important instances in the algebraic path problem, the connectivity problem, and several related problems are all solved in O(log N) time. These include the all-pair shortest paths, the minimum-weight spanning tree, the transitive closure, the connected component, the biconnected component, the articulation point, and the bridge problems, either in an undirected or a directed graph, respectively View full abstract»

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

    Page(s): 1203 - 1210
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (172 KB)  

    The Fibonacci Cube is an interconnection network that possesses many desirable properties that are important in network design and application. The Fibonacci Cube can efficiently emulate many hypercube algorithms and uses fewer links than the comparable hypercube, while its size does not increase as fast as the hypercube. However, most Fibonacci Cubes (more than 2/3 of all) are not Hamiltonian. In this paper, we propose an Extended Fibonacci Cube (EFC1) with an even number of nodes. It is defined based on the same sequence F(i)=F(i-1)+F(i-2) as the regular Fibonacci sequence; however, its initial conditions are different. We show that the Extended Fibonacci Cube includes the Fibonacci Cube as a subgraph and maintains its sparsity property. In addition, it is Hamiltonian and is better in emulating other topologies. Specifically, the Extended Fibonacci Cube can embed binary trees more efficiently than the regular Fibonacci Cube and is almost as efficient as the hypercube, even though the Extended Fibonacci Cube is a much sparser network than the hypercube. We also propose a series of Extended Fibonacci Cubes with even number of nodes. Any Extended Fibonacci Cube (EFCk, with k⩾) in the series contains the node set of any other cube that precedes EFCk in the series. We show that any Extended Fibonacci Cube maintains virtually all the desirable properties of the Fibonacci Cube. The EFCks can be considered as flexible versions of incomplete hypercubes, which eliminates their restriction on the number of nodes, and, thus, makes it possible to construct parallel machines with arbitrary sizes View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A parallel algorithm for constructing a labeled tree

    Page(s): 1236 - 1240
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (248 KB)  

    A tree T is labeled when the n vertices are distinguished from one another by names such as v1, v2…vn . Two labeled trees are considered to be distinct if they have different vertex labels even though they might be isomorphic. According to Cayley's tree formula, there are nn-2 labeled trees on n vertices. Prufer used a simple way to prove this formula and demonstrated that there exists a mapping between a labeled tree and a number sequence. From his proof, we can find a naive sequential algorithm which transfers a labeled tree to a number sequence and vice versa. However, it is hard to parallelize. In this paper, we shall propose an O(log n) time parallel algorithm for constructing a labeled tree by using O(n) processors and O(n log n) space on the EREW PRAM computational model View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Generalized algorithm for parallel sorting on product networks

    Page(s): 1211 - 1225
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (392 KB)  

    We generalize the well-known odd-even merge sorting algorithm, originally due to Batcher (1968), and show how this generalized algorithm can be applied to sorting on product networks. If G is an arbitrary factor graph with N nodes, its r-dimensional product contains Nr nodes. Our algorithm sorts Nr keys stored in the r-dimensional product of G in O(rrF(N)) time, where F(N) depends on G. We show that, for any factor graph G, F(N) is, at most, O(N), establishing an upper bound of O(r2 N) for the time complexity of sorting Nr keys on any product network. For product networks with bounded r(e.g. for grids), this leads to the asymptotic complexity of O(N) to sort Nr keys, which is optimal for several instances of product networks. There are factor graphs for which F(N)=O(log2 N), which leads to the asymptotic running time of O(log2 N) to sort Nr keys. For networks with bounded N (e.g. in the hypercube N=2, fixed), the asymptotic complexity becomes O(r2). We show how to apply the algorithm to several cases of well-known product networks, as well as others introduced recently. We compare the performance of our algorithm to well-known algorithms developed specifically for these networks, as well as others. The result of these comparisons led us to conjecture that the proposed algorithm is probably the best deterministic algorithm that can be found in terms of the low asymptotic complexity with a small constant View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Detecting corrupted pages in M replicated large files

    Page(s): 1241 - 1245
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (212 KB)  

    A file in a distributed database system is replicated on M sites and may contain corrupted pages. Abdel-Ghafiar and El Abbadi gave a detection scheme assuming that the number of corrupted pages f<M/2. We replace this assumption by a much weaker one, that, for each page, the majority of copies are correct. Our schemes are based on the structure of the Reed-Solomon code, as proposed by Abdel-Ghaffar and El Abbadi for M=2 View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Backbone networks using rotation counters

    Page(s): 1288 - 1298
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (216 KB)  

    We use simulations to evaluate the performance of an RCP ring network for network interconnection. The RCP (Rotation Counter Protocol) is a token ring protocol that has in its token a special field for counting down of the number of packets transmitted in one cycle. By correlating the queuing performance of each station through the counter, the delay of the high-priority data can be made almost independent and the delay jitters made small, Thus, RCP can be utilized in real-time control environment or integrated services environment. The performance of the backbone networks is investigated, using voice and data integration; the performance measures are mean voice delay, delay jitters, and mean data delay. Various buffering schemes for voice packets are compared. The effect of asymmetric data load is examined, and RCP is found to be capable of minimizing the loading effect of one station to the other View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A unified architecture for the computation of B-spline curves and surfaces

    Page(s): 1275 - 1287
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (428 KB)  

    B-Splines, in general, and Non-Uniform Rational B-Splines (NURBS), in particular, have become indispensable modeling primitives in computer graphics and geometric modeling applications. In this paper, a novel high-performance architecture for the computation of uniform, nonuniform, rational, and nonrational B-Spline curves and surfaces is presented. This architecture has been derived through a sequence of steps. First, a systolic architecture for the computation of the basis function values, the basis function evaluation array (the BFEA), is developed. Using the BFEA as its core, an architecture for the computation of NURBS curves is constructed. This architecture is then extended to compute NURBS surfaces. Finally, this architecture is augmented to compute the surface normals, so that the output from this architecture can be directly used for rendering the NURBS surface View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Nearly optimal one-to-many parallel routing in star networks

    Page(s): 1196 - 1202
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (220 KB)  

    Star networks were proposed recently as an attractive alternative to the well-known hypercube models for interconnection networks. Extensive research has been performed that shows that star networks are as versatile as hypercubes. This paper is an effort in the same direction. Based on the well-known paradigms, we study the one-to-many parallel routing problem on star networks and develop an improved routing algorithm that finds n-1 node-disjoint paths between one node and a set of other n-1 nodes in the n-star network. These parallel paths are proven of minimum length within a small additive constant, and the running time of our algorithm is bounded by O(n2). More specifically, given a node s and n-1 other nodes {t1, t2 , …, tn-1} in the n-star network, our algorithm constructs n-1 node-disjoint paths P1, P2, …, Pn-1, where Pi is a path from s to tj of length at most dist(s, tj)+6 and dist(s, t j) is the distance, i.e., the length of a shortest path, from s to tj, for i=1, 2, …, n-1.The best bound on the path length by previously known algorithms for the same problem is 5(n-2)≈10Δn/3, where Δn=max{dist(s, t)} is the diameter of the n-star network 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

    Page(s): 1268 - 1274
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (220 KB)  

    In a distributed environment, tasks often have processing demands at multiple different sites. A distributed task is usually divided into several subtasks, each to be executed in order at some site. In a real-time system, an overall deadline is usually specified by an application designer indicating when a distributed task is to be finished. In this paper, we present and analyze techniques for automatically translating the overall deadline into deadlines for the individual subtasks View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Fault-tolerant ring embedding in a star graph with both link and node failures

    Page(s): 1185 - 1195
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (364 KB)  

    The star graph interconnection network has been recognized as an attractive alternative to the hypercube network. Previously, the star graph has been shown to contain a Hamiltonian cycle. In this paper, we consider an injured star graph with some faulty links and nodes. We show that even with fe⩽n-3 faulty links, a Hamiltonian cycle still can be found in an n-star, and that with fv⩽n-3 faulty nodes, a ring containing at most 4fv nodes less than that in a Hamiltonian cycle can be found (i.e. the ring contains at least n!-4fv nodes). In general, in an n-star with fe faulty links and fv faulty nodes, where fe+fv⩽n-3, our embedding is able to establish a ring containing at least n!-4fv nodes View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Scheduling data-flow graphs via retiming and unfolding

    Page(s): 1259 - 1267
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (256 KB)  

    Loop scheduling is an important problem in parallel processing. The retiming technique reorganizes an iteration; the unfolding technique schedules several iterations together. We combine these two techniques to obtain a static schedule with a reduced average computation time per iteration. We first prove that the order of retiming and unfolding is immaterial for scheduling a data-flow graph (DFG). From this nice property, we present a polynomial-time algorithm on the original DFG, before unfolding, to find the minimum-rate static schedule for a given unfolding factor. For the case of a unit-time DFG, efficient checking and retiming algorithms are presented View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An effective processor allocation strategy for multiprogrammed shared-memory multiprocessors

    Page(s): 1246 - 1258
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (432 KB)  

    Existing techniques for sharing the processing resources in multiprogrammed shared-memory multiprocessors, such as time-sharing, space-sharing, and gang-scheduling, typically sacrifice the performance of individual parallel applications to improve overall system utilization. We present a new processor allocation technique called Loop-Level Process Control (LLPC) that dynamically adjusts the number of processors an application is allowed to use for the execution of each parallel section of code, based on the current system load. This approach exploits the maximum parallelism possible for each application without overloading the system. We implement our scheme on a Silicon Graphics Challenge multiprocessor system and evaluate its performance using applications from the Perfect Club benchmark suite and synthetic benchmarks. Our approach shows significant improvements over traditional time-sharing and gang-scheduling. It has performance comparable to, or slightly better than, static space-sharing, but our strategy is more robust since, unlike static space-sharing, it does not require a priori knowledge of the applications' parallelism characteristics 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.

Full Aims & Scope

Meet Our Editors

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