Parallel Computer

In the paper, parallelization of finite element modeling of solidification is considered. The core of this modeling is solving large sparse linear systems. The Aztec library is used for implementing the model problem on massively parallel computers. Now the complete parallel code is available. The performance results of numerical experiments carried out on the IBM SP2 parallel computer are presented.

Simulation of quantum computers using classical computers is a hard problem with high memory and computational requirements. Parallelization can alleviate this problem, allowing the simulation of more qubits at the same time or the same number of qubits to be simulated in less time. A promising approach is to exploit the high performance computing capabilities provided by the latest graphical processing units. In this paper we present a parallel implementation of the QC-lib quantum computer simulator on the GPU using the CUDA programming model. The proposed scheme for partitioning the terms that describe the state of a quantum register takes advantage of the specific characteristics of the CUDA memory spaces and allows for an efficient parallelization of the general singe qubit operator. Experimental results indicate that very good speed-ups can be obtained in contrast with the sequential implementation.

Porting well known computer vision algorithms to low power, high performance computing devices such as SIMD linear processor arrays can be a challenging task. One especially useful such algorithm is the color-based particle filter, which has been applied successfully by many research groups to the problem of tracking non-rigid objects. In this paper, we propose an implementation of the color-based particle filter suitable for SIMD processors. The main focus of our work is on the parallel computation of the particle weights. This step is the major bottleneck of standard implementations of the color-based particle filter since it requires the knowledge of the histograms of the regions surrounding each hypothesized target position. We expect this approach to perform faster in an SIMD processor than an implementation in a standard desktop computer even running at much lower clock speeds.

At Queen Mary, University of London, we have over twenty years of experience in Parallel Computing Applications, mostly on "massively parallel systems", such as the Distributed Array Processors (DAPs). The applications in which we were involved included design of numerical subroutine libraries, Finite Element software, graphics tools, the physics of organic materials, medical imaging, computer vision and more recently, Financial modelling. Two of the projects related to the latter are described in this paper, namely Portfolio Optimisation and Financial Risk Assessment.

This paper describes our experience with developing a parallel weighted- least-square (WLS) state estimation (SE) program for shared-memory parallel computers. Since the key computational kernel of the WLS algorithm based on the Newton-Raphson approach is a solver of sparse linear equations, a significant part of our effort was focused on selecting, implementing and evaluating this algorithm. An optimized shared memory version of the conjugate gradient (CG) algorithm was found to be competitive to state-of-the-art implementation of LU solvers for the SE problem on the SGI Altix, a shared memory architecture. We also ported the full SE algorithm including CG to the Cray MTA-2 shared memory multithreaded architecture and investigated its performance.

Abstract. The Lanczos algorithm is a very effective method for finding extreme eigenvalues of symmetric matrices. In this paper, we present our parallel version of the Lanczos method for symmetric generalized eigenvalue problem, PLANSO. PLANSO is based on a sequential package ...

Over the past four years there has been a tremendous resurgence of "public" interest inside and outside of IBM in APL and especially APL2 as a result of 1. an awareness of the importance of for technical computing, rapid prototyping, array algorithms for parallel computing, and advanced decision support 2. state-of-the-art work being done with APL in such diverse areas as medical imaging, finance, manufacturing scheduling, spot market oil buying and selling, computer design, chip design, mathematics, knowledge based systems statistics, etc. 3. IBM establishing APL2 as its strategic interactive language and making it a key player in its technical computing strategy Due to this renewed interest, I have compiled a list of some APL-related articles on Modeling and on Knowledge-Based Expert Systems .(KBES).

... This proposed algorithm has the advantages of both the sequential FCM and parallelFCM for the clustering process in the segmentation techniques. This algorithm is very fast when the image size is large and it requires less execution time. ...

Church’s and Turing’s theses assert dogmatically that an informal notion of effective calculability is captured adequately by a particular mathematical concept of computabilty. I present analyses of calculability that are embedded in a rich historical and philosophical context, lead to precise concepts, and dispense with theses. To investigate effective calculability is to analyze processes that can in principle be carried out by calculators. This is a philosophical lesson we owe to Turing. Drawing on that lesson and recasting work of Gandy, I formulate boundedness and locality conditions for two types of calculators, namely, human computing agents and mechanical computing devices (or discrete machines). The distinctive feature of the latter is that they can carry out parallel computations. Representing human and machine computations by discrete dynamical systems, the boundedness and locality conditions can be captured through axioms for Turing computors and Gandy machines; models o...

The Choices operating system architecture [3, 4, 15] uses class hierarchies and object-oriented programming to facilitate the construction of customized operating systems for shared memory and networked multiprocessors. The software is being used in the Tapestry Parallel Computing Laboratory at the University of Illinois to study the performance of algorithms, mechanisms, and policies for parallel systems. This paper describes the architectural design and class hierarchy of the Choices memory and secondary storage management system. The mechanisms and policies of a virtual memory system implement a memory hierarchy that exploits the trade-offs between response times and storage capacities. In Choices , the notion of a memory hierarchy is represented by layers in which abstract classes define interfaces between and internal to the layers. Concrete subclasses implement new algorithms or data structures or specializations of existing ones. This paper describes the motivation for an obj...