并行算法

In the realm of sequential computing Von Neumann machine has successfully provided an unifying and universal model of computation.

并行计算模型为并行算法和并行计算机系统结构的分析与设计提供了具有指导意义的理论界面和模型框架，它是并行计算研究的重要领域。

By using the tensor product of matrix, stride permutation matrix and matrix power series under the meaning of matrix direct sum, the parallel algorithm, including the scrambling algorithm, can be expressed integrally. This description provides a convenient tool for the study and application of parallel FFT algorithm.

本文提出了矩阵直和意义下幂级数的矩阵运算，同时利用迭代形式的矩阵方程将混合基FFT的并行算法及其数码倒置算法完整地表示出来，为并行FFT算法的研究和应用提供了方便的表示手段。

This paper presents the definition of fault tolerant parallel algorithm.

给出了容错并行算法的定义，提出了一种新的基于并行复算的容错并行算法。

Parallel algorithm including parallel and distributed computation algorithm design, analysis and realization; Working section, duty level, sub- duty level, circulation level and instruction level parallel optimized translation.

并行算法包括并行和分布计算算法的设计、分析和实现；作业级、任务级、子任务级、循环级和指令级的并行优化编译。

In this paper, we present a class of equation model for image denoising and edge enhancement. The model is based on robust statistics error norm and variational theory.

本文基于稳健统计误差模函数及变分原理构造了一类用于图像去噪、增强边缘的方程模型，对相关参数进行了分析，设计了该方程有限差分数值求解的并行算法，并对分块模型及实现过程进行改进，提高了并行效率，在此基础上完成了并行数值实验，并对结果就误差模选取、并行效率等方面进行了分析。

BiConjugate Gradients Stabilized method is applied for solving large linear equation set. That the calculation results of PC modeling program match well with that of parallel algorithm shows high reliability of the latter. The parallel algorithm has established the basis for the inversion 3D Borehole-Surface electromagnetic data and also provided the tech support for the utilization of the program.

在求解大型线性方程组时，采用的是稳定的双共轭梯度算法，单机正演程序与并行正演程序的计算结果有着很好的一致性，这说明了并行算法的正确性，为进一步开发井地电法三维反演并行程序奠定了基础，也为井地电法三维正、反演程序实用化提供了技术支持。

In Chapter 4, a set of distributed graphics algorithms (including fractal image creation, Delaunay triangulation of scalled points, ray tracing and volume rendering) are presented.

在第四章介绍用DGPSE设计的一组分布式图形算法（包括分形图象，Delaunay三角剖分、光线跟踪和辐射度绘制）和分布式图形库DGPSL，并行算法涉及到对象空间并行/图象空间并行方法。

In this paper, we also discuss the theory of the parallel simplex method. Undej the 2-dimensional space case, we present that simplex seriary have some properties, but we adopt some repairing measures. We proof the convergence of the parallel algorithm. At last, We give some numerical examples to indicate that the new idea is sometimes useful and helpful.

本文也对算法做了一些理论分析，主要在二维情形下，讨论了算法形成的单纯形序列本身所具有的性质，并对其做了相应的补救措施，对算法的收敛性给予了证明，最后做了一些数值实验，由于软硬件条件的限制，并行算法未能在并行计算机上实现，鉴于这种情况，我们所做的数值实验均是在串行机上完成的。

We found that undernatural number coding the population size can be very small,about 6 to 10 is enough,which isvery different from the classic GA.

实验结果表明，这种并行算法实现简单，效果也不错，而且从概率算法的角度来说，这种并行算法的加速比可以认为是线性的

Then a new method named parallel recomputing based fault tolerant parallel algorithm is proposed. The authors present the design of PRBFTPA for matrix LU decomposition, and evaluate them by experiments on a cluster system and compare them with the method of checkpointing.

针对许多计算密集型任务中的矩阵LU分解设计了相应的基于并行复算的容错并行算法，并对设计的矩阵LU分解的容错并行算法的性能进行了评估并与checkpointing方法进行了对比。

The split between the two groups can hardly be papered over.

这两个团体间的分歧难以掩饰。

This approach not only encourages a greater number of responses, but minimizes the likelihood of stale groupthink.

这种做法不仅鼓励了更多的反应，而且减少跟风的可能性。