Efficient Parallel Implementation of a Compact Higher-Order Maxwell Solver Using Spatial Filtering

2000 ◽  
pp. 75-87
Author(s):  
Ramesh K. Agarwal
Keyword(s):  
Parallel Implementation ◽  
Spatial Filtering ◽  
Higher Order

$\mathcal{HDC}$: A HIGHER-ORDER LANGUAGE FOR DIVIDE-AND-CONQUER

2000 ◽  
Vol 10 (02n03) ◽  
pp. 239-250 ◽  
Author(s):  
CHRISTOPH A. HERRMANN ◽  
CHRISTIAN LENGAUER
Keyword(s):  
Parallel Programming ◽  
Parallel Implementation ◽  
Higher Order ◽  
Divide And Conquer ◽  
Functional Language ◽  
Polynomial Multiplication ◽  
Functional Program ◽  
Design Decisions ◽  
Order Language

We propose the higher-order functional style for the parallel programming of algorithms. The functional language [Formula: see text], a subset of the language Haskell, facilitates the clean integration of skeletons into a functional program. Skeletons are predefined programming schemata with an efficient parallel implementation. We report on our compiler, which translates [Formula: see text] programs into C+MPI, especially on the design decisions we made. Two small examples, the n queens problem and Karatsuba's polynomial multiplication, are presented to demonstrate the programming comfort and the speedup one can obtain.

A Theoretical Framework for Parallel Implementation of Deep Higher Order Neural Networks

2016 ◽  
pp. 351-361
Author(s):  
Shuxiang Xu ◽  
Yunling Liu
Keyword(s):  
Neural Networks ◽  
Distributed System ◽  
Learning Algorithm ◽  
Parallel Implementation ◽  
Theoretical Framework ◽  
Higher Order ◽  
Area Network ◽  
System Environment ◽  
New Learning ◽  
Higher Order Neural Networks

This chapter proposes a theoretical framework for parallel implementation of Deep Higher Order Neural Networks (HONNs). First, we develop a new partitioning approach for mapping HONNs to individual computers within a master-slave distributed system (a local area network). This will allow us to use a network of computers (rather than a single computer) to train a HONN to drastically increase its learning speed: all of the computers will be running the HONN simultaneously (parallel implementation). Next, we develop a new learning algorithm so that it can be used for HONN learning in a distributed system environment. Finally, we propose to improve the generalisation ability of the new learning algorithm as used in a distributed system environment. Theoretical analysis of the proposal is thoroughly conducted to verify the soundness of the new approach. Experiments will be performed to test the new algorithm in the future.

scholarly journals RCS Computation by Parallel MoM Using Higher-Order Basis Functions

2012 ◽  
Vol 2012 ◽  
pp. 1-8 ◽  
Author(s):  
Ying Yan ◽  
Yu Zhang ◽  
Chang-Hong Liang ◽  
Hui Zhao ◽  
D. García-Doñoro
Keyword(s):  
Integral Equation ◽  
Cross Section ◽  
Method Of Moments ◽  
Load Balance ◽  
Message Passing ◽  
Message Passing Interface ◽  
Parallel Implementation ◽  
Radar Cross Section ◽  
Higher Order ◽  
Basis Functions

A Message-Passing Interface (MPI) parallel implementation of an integral equation solver that uses the Method of Moments (MoM) with higher-order basis functions has been proposed to compute the Radar Cross-Section (RCS) of various targets. The block-partitioned scheme for the large dense MoM matrix is designed to achieve excellent load balance and high parallel efficiency. Some numerical results demonstrate that higher-order basis in this parallelized scheme is more efficient than the conventional RWG method and able to efficiently analyze RCS of various electrically large platforms.

scholarly journals A parallel implementation of sediment transport and bottom surface reconstruction problems on the basis of higher-order difference schemes

2015 ◽  
pp. 256-267
Author(s):  
А.И. Сухинов ◽  
А.Е. Чистяков ◽  
А.А. Семенякина ◽  
А.В. Никитина
Keyword(s):  
Sediment Transport ◽  
Parallel Implementation ◽  
Bottom Topography ◽  
Diffusion Problem ◽  
Higher Order ◽  
Difference Schemes ◽  
Bottom Surface ◽  
Multiprocessor Computer ◽  
Order Difference ◽  
Convection Diffusion

Статья посвящена изучению дискретных аналогов операторов конвективного и диффузионного переносов четвертого порядка точности в случае частичной заполненности ячеек. Выполнено сопоставление результатов расчета задачи транспорта веществ на основе схем второго и четвертого порядков точности. Из сопоставления результатов численных экспериментов следует, что для задачи диффузии удалось повысить точность в 66.7 раз, а для задачи диффузии-конвекции в 48.7 раз. Для решения двумерной задачи диффузии-конвекции на основе схем повышенного порядка точности была построена библиотека двухслойных итерационных методов, предназначенных для решения девятидиагональных сеточных уравнений на многопроцессорной вычислительной системе. Предложен алгоритм, предназначенный для восстановления рельефа дна акватории на основе гидрографической информации (глубины водоема в отдельных точках или изолиний уровня), и выполнена его численная реализация. На основе полученного метода решения задачи построена карта рельефа дна Азовского моря. Discrete analogs of convective and diffusive transport operators of the fourth order of accuracy are studied in the case of partially filled cells. The numerical results obtained when solving the sediment transport problem on the basis of difference schemes of the second and fourth orders of accuracy are compared. These results show that the accuracy of the solutions to the diffusion problem and the convection-diffusion problem increases by a factor of 66.7 and 48.7, respectively. A library of two-layer iterative methods is built to solve the two-dimensional convection-diffusion problem on the basis of higher-order schemes for nine-diagonal difference equations on a multiprocessor computer system. An algorithm is proposed to reconstruct the submarine bottom topography on the basis of hydrographic information (the water depth at a number of points or contour levels) and its numerical implementation is performed. The proposed method is used to draw a map of the bottom relief of the Azov sea.

A Theoretical Framework for Parallel Implementation of Deep Higher Order Neural Networks

2016 ◽  
pp. 1-11
Author(s):  
Shuxiang Xu ◽  
Yunling Liu
Keyword(s):  
Neural Networks ◽  
Distributed System ◽  
Learning Algorithm ◽  
Parallel Implementation ◽  
Theoretical Framework ◽  
Higher Order ◽  
Area Network ◽  
System Environment ◽  
New Learning ◽  
Higher Order Neural Networks

This chapter proposes a theoretical framework for parallel implementation of Deep Higher Order Neural Networks (HONNs). First, we develop a new partitioning approach for mapping HONNs to individual computers within a master-slave distributed system (a local area network). This will allow us to use a network of computers (rather than a single computer) to train a HONN to drastically increase its learning speed: all of the computers will be running the HONN simultaneously (parallel implementation). Next, we develop a new learning algorithm so that it can be used for HONN learning in a distributed system environment. Finally, we propose to improve the generalisation ability of the new learning algorithm as used in a distributed system environment. Theoretical analysis of the proposal is thoroughly conducted to verify the soundness of the new approach. Experiments will be performed to test the new algorithm in the future.

Sign in / Sign up

Export Citation Format

Share Document