Multiprocessor Computer Research Articles

Technology of high-performance image processing on multiprocessor computer

An experimental framework SSCCIP for high-performance image processing on multiprocessor computer is described. The basic principles of the system construction, various architectural solutions, and an example of implementing a specific technology of the image processing are discussed.

Study of the anomalous magnetic field structure in the Ural region using parallel algorithms

The anomalous structure in the magnetic field of the Ural Region has been studied in the segment bounded by 52°–64° N and 54°–66° E. Analytical apparatus for upward continuation of airborne magnetic data to different heights was applied. To recalculate magnetic field, parallel algorithms and software for multiprocessor computers were used. Maps of magnetic anomalies for different ranges of wave lengths showing the distribution of magnetization in the layers of the Earth’s crust were built.

CaPSID: A bioinformatics platform for computational pathogen sequence identification in human genomes and transcriptomes

BackgroundIt is now well established that nearly 20% of human cancers are caused by infectious agents, and the list of human oncogenic pathogens will grow in the future for a variety of cancer types. Whole tumor transcriptome and genome sequencing by next-generation sequencing technologies presents an unparalleled opportunity for pathogen detection and discovery in human tissues but requires development of new genome-wide bioinformatics tools.ResultsHere we present CaPSID (Computational Pathogen Sequence IDentification), a comprehensive bioinformatics platform for identifying, querying and visualizing both exogenous and endogenous pathogen nucleotide sequences in tumor genomes and transcriptomes. CaPSID includes a scalable, high performance database for data storage and a web application that integrates the genome browser JBrowse. CaPSID also provides useful metrics for sequence analysis of pre-aligned BAM files, such as gene and genome coverage, and is optimized to run efficiently on multiprocessor computers with low memory usage.ConclusionsTo demonstrate the usefulness and efficiency of CaPSID, we carried out a comprehensive analysis of both a simulated dataset and transcriptome samples from ovarian cancer. CaPSID correctly identified all of the human and pathogen sequences in the simulated dataset, while in the ovarian dataset CaPSID’s predictions were successfully validated in vitro.

Numerical analysis of a model of metal solidification, 2D case

The work discusses numerical simulation of metal solidification. The difficulties of calculatibg such processes are that simulation is conducted at several scale levels simultaneously. At present experimental studies cover numerous aspects of metal solidification; however, there is no generalized concept of this process. The model applied in this study is based on the concept of solidified alloy in the form of a porous medium, where propagation of perturbations is described by the Biot equations. In order to describe nucleation, the modified Cahn-Hillard equation is applied. Previously a one-dimensional (1D) numerical model has been developed, its convergence has been demonstrated and a possibility of obtaining of various modes of solidification has been shown at modification of parameters. This work presents the 2D numerical model and examples of calculations. Since a multiscale approach requires high CPU resources, the 2D calculations are based on explicit and explicit-iterative algorithms which are efficiently implemented using a multiprocessor computer.

New Approach Construction for Wireless ZigBee Sensor Based on Embedding Pancake Graphs

Wireless Sensor Networks (WSN) based on the IEEE 802.15.4 standard are constantly expanding. Applications like production control, building control are more and more based on WSN because of their energy efficiency, self organizing capacity and protocol flexibility. However, the construction of Cluster-Tree networks based on the beacon mode Pancake graphs is still undefined by the IEEE 802.15.4 standard. In order to enable the construction of such topology, i.e., Beacon Cluster-Tree based on Pancake graphs, we present, in this paper, a new topology construction approach. The Pancake is one of the Cayley graphs that were proposed as alternative to the Hypercube for interconnecting processors in parallel computers. This network offers attractive and desirable properties: Vertex symmetry, small degree and diameter, extensibility, high connectivity, easy routing, regularity of topology, fault-tolerance, and embed ability of other topologies. We present in this work the many-to-one embedding of Multiply-Twisted Hypercube into the Pancake networks with dilation 5 as a new approach for wireless networks. The presented approach is based on the exploitation of RF front-end capabilities in treating multipath signals and, thus, avoiding the introduction of beacon or Super Frames scheduling algorithms. Avoiding the introduction of scheduling algorithms ensures a simple solution that could be easily implemented and executed by ZigBee sensor nodes.

Heat transfer in a three-dimensional rectangular cavity oriented at an angle to the free-stream

We consider a flow of a viscous incompressible heat-conducting fluid over a cubic cavity. The heat transfer on the bottom of the cavity rotated at an angle to the free stream is studied numerically. The numerical algorithm includes a finite-difference approximation in the spatial coordinates, a semi-implicit time integration method, and a modified version of an iterative stabilized method of biconjugate gradients with an algebraic multigrid preconditioner for solving the Poisson equation. The algorithm is designed for the use of multiprocessor computers. Two different inlet flow conditions are considered: a steady-state flow and a steady flow with superimposed periodic perturbations. In the first case, it is shown that the integral heat transfer rate increases monotonically with increase in the cavity rotation angle α. For α = 45°, the increase in the heat flux amounts to 20%. The presence of periodic disturbances may result in up to 3-fold increase in the integral heat transfer rate as compared to the case of the steady-state inlet flow. The enhancement of heat transfer occurs when the frequency of the inlet flow disturbances is close to the frequency of unstable modes of the mixing layer formed at the upper boundary of the cavity.

Parallel computing in forest fires two-dimension modeling

A two-dimensional two-phase mathematical model of forest fires is considered. The model is derived by averaging the three-dimensional equations over the thickness of the forest fuel material layer. The results of a demonstrative numerical simulation of the process of forest fire spread using multiprocessor computer are presented.

Real-time simulation model development of single crystalline photovoltaic panels using fast computation methods

Real-time simulation and rapid prototyping of power electronics, critical loads, and control systems have prompted recent interest in accurate electrical terminal models of photovoltaic (PV) panels and array systems. Advancement in computing technologies, such as parallel computing and digital signal processing techniques for real-time simulations have allowed the prototyping of novel apparatus to be investigated in a virtual system under a wide range of realistic conditions repeatedly, safely, and economically. This paper accesses numerical iteration methods, selects appropriate techniques, and combines them with model construction methods well suited for boosting the computational speed of an electro-thermal dynamic model of a PV panel. The target computational engine is a parallel processor based real-time simulator to be used in a power hardware-in-the-loop (PHIL) application. Significant improvements resulting from the proposed modeling approach in computation time and numerical convergence speed are verified using experimental results for the target PV panel using Opal-RT’s RT-Lab Matlab/Simulink based real-time engineering simulator.

Parallel Logic Simulation: Myth or Reality?

The rapid adoption of multiprocessor computers creates a perfect environment for parallel EDA algorithms. Among various EDA applications, parallel logic simulation seems the most promising. As processors become faster and designs grow larger, better performance might well be expected from parallel simulation. However, this is typically not the case. Speedup is difficult to achieve, and pitfalls in evaluating simulation speedup can offer false promises. Is parallel logic simulation a myth or reality? In this article, the authors answer this question from the perspective of an industrial parallel simulation vendor.

From Wheatstone to Cameron and beyond: overview in 3-D and 4-D imaging technology

This paper reviews three-dimensional (3-D) and four-dimensional (4-D) imaging technology, from Wheatstone through today, with some prognostications for near future applications. This field is rich in variety, subject specialty, and applications. A major trend, multi-view stereoscopy, is moving the field forward to real-time wide-angle 3-D reconstruction as breakthroughs in parallel processing and multi-processor computers enable very fast processing. Real-time holography meets 4-D imaging reconstruction at the goal of achieving real-time, interactive, 3-D imaging. Applications to telesurgery and telemedicine as well as to the needs of the defense and intelligence communities are also discussed.

Injection of the maximal payload under random perturbations of parameters

Consideration was given to determination of the maximal mass of the payload that can be inserted into an elliptical orbit by the "SOYUZ-2" launch vehicle with a probability not smaller than the predefined threshold and to the case where the launch vehicle parameters and the atmospheric characteristics are described by normally distributed random variables with certain expectations and rms deviations. A procedure was proposed to estimate the payload mass, and the results of numerical modeling on the multiprocessor computer system using the real data were presented.

Eddy-resolving 1/10° model of the World Ocean

The first results on simulating the intra-annual variability of the World Ocean circulation by use of the eddy-resolving model are considered. For this purpose, a model of the World Ocean with a 1/10° horizontal resolution and 49 vertical levels was developed (a 1/10 × 1/10 × 49 model of the World Ocean). This model is based on the traditional system of three-dimensional equations of the large-scale dynamics of the ocean and boundary conditions with an explicit allowance for water fluxes on the free surface of the ocean. The equations are written in the tripolar coordinate system. The numerical method is based on the separation of the barotropic and baroclinic components of the solution. Discretization in time is implemented using explicit schemes allowing effective parallelization for a large number of processors. The model uses the sub-models of the boundary layer of the atmosphere and the submodel of sea-ice thermodynamics. The model of the World Ocean was developed at the Institute of Numerical Mathematics of the Russian Academy of Sciences (INM RAS) and the P.P. Shirshov Institute of Oceanogy (IO RAS). The formulation of the problem of simulating the intra-annual variability of thermohydrodynamic processes of the World Ocean and the parameterizations that were used are considered. In the numerical experiment, the temporal evolution of the atmospheric effect is determined by the normal annual cycle according to the conditions of the international Coordinated Ocean-Ice Reference Experiment (CORE-I). The calculation was carried out on a multiprocessor computer with distributed memory; 1601 computational cores were used. The presented analysis demonstrates that the obtained results are quite satisfactory when compared to the results that were obtained by other eddy-resolving models of the global ocean. The analysis of the model solution is, to a larger extent, of a descriptive character. A detailed analysis of the results is to be presented in following works. This experiment is a significant first step in developing the eddy-resolving model of the World Ocean.

Using the FEM Meshes Adaption and Genetic Algorithms for Identification of Permeability in Normal Direction of Anisotropic Sheets

This paper proposes the identification of the B-H curve of anisotropic magnetic laminations in the direction normal to the sheet surface. An automated procedure for linking an identification algorithm implemented in a Matlab environment with a commercial finite-element code Opera for 3-D magnetic field analysis has been used. In particular, an optimization procedure is proposed, where the computation time is reduced by adjusting the refinement of the FEM mesh. The use of a multiprocessor computer allowed to perform parallel computations and to realize the calculation in a reasonable time. The finite element mesh density is changed during operation of the procedure depending on the level of magnetic flux density difference between calculation and measurements.

Modeling of III-D problems of gas dynamics on multiprocessing computers and GPU

The paper deals with a parallel algorithm for calculations on multiprocessor computers and GPU accelerators. The calculations of shock waves interaction with low density bubble results are presented [1]. This algorithm combines a possibility to capture a high resolution of shock waves, the second-order accuracy for TVD schemes, and a possibility to observe a low-level diffusion of the advection scheme.

CELLS v1.0: updated and parallelized version of an electrical scheme to simulate multiple electrified clouds and flashes over large domains

Abstract. The paper describes the fully parallelized electrical scheme CELLS which is suitable to simulate explicitly electrified storm systems on parallel computers. Our motivation here is to show that a cloud electricity scheme can be developed for use on large grids with complex terrain. Large computational domains are needed to perform real case meteorological simulations with many independent convective cells. The scheme computes the bulk electric charge attached to each cloud particle and hydrometeor. Positive and negative ions are also taken into account. Several parametrizations of the dominant non-inductive charging process are included and an inductive charging process as well. The electric field is obtained by inverting the Gauss equation with an extension to terrain-following coordinates. The new feature concerns the lightning flash scheme which is a simplified version of an older detailed sequential scheme. Flashes are composed of a bidirectional leader phase (vertical extension from the triggering point) and a phase obeying a fractal law (with horizontal extension on electrically charged zones). The originality of the scheme lies in the way the branching phase is treated to get a parallel code. The complete electrification scheme is tested for the 10 July 1996 STERAO case and for the 21 July 1998 EULINOX case. Flash characteristics are analysed in detail and additional sensitivity experiments are performed for the STERAO case. Although the simulations were run for flat terrain conditions, they show that the model behaves well on multiprocessor computers. This opens a wide area of application for this electrical scheme with the next objective of running real meterological case on large domains.

Embedding n-Dimensional Crossed Hypercube into Pancake Graphs

Among Cayley graphs on the symmetric group, the pancake graph is one as a viable interconnection scheme for parallel computers, which has been examined by a number of researchers. The pancake was proposed as alternatives to the hypercube for interconnecting processors in parallel computers. Some good and attractive properties of this interconnection network include: vertex symmetry, small degree, a sub-logarithmic diameter, extendability, and high connectivity (robustness), easy routing, and regularity of topology, fault tolerance, extensibility and embeddability of other topologies. In this paper, we present the many-to-one dilation 5 embedding of n-dimensional crossed hypercube into n-dimensional pancake patients. These predictors, however, need further work to validate reliability.

Designing a Multicore and Multiprocessor Individual-Based Simulation Engine

This article describes the design of an individual-based simulation engine that can harness the full potential of modern general-purpose multicore and multiprocessor computers. This design aims to enable interactive simulations of highly dynamic multiagent systems in which entities can move, change, appear, disappear, and interact with one another and the user at any time.

A Novel College Network Resource Management Method using Cloud Computing

Abstract At present information construction of college mainly has construction of college networks and management information system; there are many problems during the process of information. Cloud computing is development of distributed processing, parallel processing and grid computing, which make data stored on the cloud, make software and services placed in the cloud and build on top of various standards and protocols, you can get it through all kinds of equipments. This article introduces cloud computing and function of cloud computing, then analyzes the exiting problems of college network resource management, the cloud computing technology and methods are applied in the construction of college information sharing platform.

An efficient energy and schedule length model for multiprocessor computers

Multiprocessors have emerged as a powerful computing means for running real-time applications, especially where a uniprocessor system would not be sufficient to execute all the tasks. Such a computing environment requires an efficient algorithm to determine when and on which processor a given task should be executed. An efficient scheduling of parallel tasks onto the multiprocessors is known to be an NP-hard problem. This paper focuses on the combinational optimisation problem, namely the problem of minimising schedule length with energy consumption constraint and the problem of minimising energy consumption with schedule length constraint for independent parallel tasks. These problems emphasise the trade-off between power and performance and are defined such that the power-performance product is optimised by fixing one factor and minimising other and vice versa. The performance of the proposed algorithm with optimal solution is validated analytically and compared with Particle Swarm Optimisation (PSO) and Genetic Algorithm (GA).

Post-Editing Error Correction Algorithm For Speech Recognition using Bing Spelling Suggestion

ASR short for Automatic Speech Recognition is the process of converting a spoken speech into text that can be manipulated by a computer. Although ASR has several applications, it is still erroneous and imprecise especially if used in a harsh surrounding wherein the input speech is of low quality. This paper proposes a post-editing ASR error correction method and algorithm based on Bing’s online spelling suggestion. In this approach, the ASR recognized output text is spell-checked using Bing’s spelling suggestion technology to detect and correct misrecognized words. More specifically, the proposed algorithm breaks down the ASR output text into several word-tokens that are submitted as search queries to Bing search engine. A returned spelling suggestion implies that a query is misspelled; and thus it is replaced by the suggested correction; otherwise, no correction is performed and the algorithm continues with the next token until all tokens get validated. Experiments carried out on various speeches in different languages indicated a successful decrease in the number of ASR errors and an improvement in the overall error correction rate. Future research can improve upon the proposed algorithm so much so that it can be parallelized to take advantage of multiprocessor computers.