Message-passing Multiprocessor Systems Research Articles

A parallel text document clustering algorithm based on neighbors

In this paper, we propose a new parallel algorithm for text document clustering based on the concept of neighbor (Guha et al. in Inf Syst 25(5):345---366, 2000). If two documents are similar enough, they are considered as neighbors of each other. The new algorithm is named parallel k-means based on neighbors (PKBN), and it is a parallel version of sequential k-means based on neighbors (SKBN) that we proposed in Luo et al. (Data Knowl Eng 68(11):1271---1288, 2009). PKBN fully exploits the data-parallelism of SKBN and adopts a new parallel pair-generating method to build the neighbor matrix. Our new parallel pair-generating method causes less communication overhead between processors than existing methods. PKBN is designed for message-passing multiprocessor systems and is implemented on a cluster of Linux workstations to analyze its performance. Our experimental results on real-life data sets demonstrate that PKBN is very efficient and has good scalability with respect to the number of processors and the size of data set.

Parallel bisecting k-means with prediction clustering algorithm

In this paper, we propose a new parallel clustering algorithm, named Parallel Bisecting k -means with Prediction (PBKP), for message-passing multiprocessor systems. Bisecting k -means tends to prod...

A parallel asynchronous implementation of the ϵ-relaxation method for the linear minimum cost flow problem

In this paper we present a parallel asynchronous implementation of the ϵ-relaxation method for solving the linear minimum cost flow problem on distributed memory message-passing multiprocessor systems. The general structure of the method is well suited to efficient parallelization, since a single iteration can be performed on several nodes simultaneously. We describe the implementation details of the parallel version on both a Fujitsu AP1000 and a cluster of Digital Alpha workstations connected via FDDI links. The results obtained demonstrate that our implementation is capable of substantial speedups.

Partitioning and mapping nested loops on multiprocessor systems

A method for executing nested loops with constant loop-carried dependencies in parallel on message-passing multiprocessor systems to reduce communication overhead is presented. In the partitioning phase, the nested loop is divided into blocks that reduce the interblock communication, without regard to the machine topology. The execution ordering of the iterations is defined by a given time function based on L. Lamport's (1974) hyperplane method. The iterations are then partitioned into blocks so that the execution ordering is not disturbed, and the amount of interblock communication is minimized. In the mapping phase, the partitioned blocks are mapped onto a fixed-size multiprocessor system in such a manner that the blocks that have to exchange data frequently are allocated to the same processor or neighboring processors. A heuristic mapping algorithm for hypercube machines is proposed.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Parallel rational interpolation

A fast parallel algorithm for rational interpolation based on orthogonal polynomials, which is suitable for both shared-memory and message-passing multiprocessor systems is proposed. In the shared-memory case with N + 1 identical processors, the algorithm requires 0(N log N) parallel arithmetic steps to construct all rational interpolants at once, where N +1 is the number of data points. Extensions to message-passing multiprocessor systems such as the hypercube are also discussed. The hypercube version of the algorithm requires 0(N log N) inter-processor communication overhead. Thus in effect, the algorithm constructs each rational interpolant using O(log N) parallel arithmetic and O(log N) communication steps.

A parallel algorithm for sparse symbolic Cholesky factorization on a multiprocessor

We develop an algorithm for computing the symbolic Cholesky factorization of a large sparse symmetric positive definite matrix. The algorithm is intended for a message-passing multiprocessor system, such as the hypercube, and is based on the concept of elimination forest. In addition, we provide an algorithm for computing these forests along with a discussion of the algorithm's complexity and a proof of its correctness.

Using data partitioning to implement a parallel assembler

A technique for implementing algorithms on a multiprocessor computer system is data partitioning, in which input data for a problem is partitioned among many processors that cooperate to solve the problem. This paper demonstrates that data partitioning is a good method for implementing an assembler on a message-passing multiprocessor system: it yields a speedup exceeding a factor of 6 with eight processors. We compare several alternative methods for distributing program text and sharing global information among the processors executing the assembler: operations that are important for a variety of applications implemented with data partitioning.

A parallel graph partitioning algorithm for a message-passing multiprocessor

We develop a parallel algorithm for partitioning the vertices of a graph intop≥2 sets in such a way that few edges connect vertices in different sets. The algorithm is intended for a message-passing multiprocessor system, such as the hypercube, and is based on the Kernighan-Lin algorithm for finding small edge separators on a single processor.(1) We use this parallel partitioning algorithm to find orderings for factoring large sparse symmetric positive definite matrices. These orderings not only reduce fill, but also result in good processor utilization and low communication overhead during the factorization. We provide a complexity analysis of the algorithm, as well as some numerical results from an Intel hypercube and a hypercube simulator.