1

适合于新手学习并行化MPI程序以及并行思想学习(This is a parallel program learning material for beginners.)

基于fpga的矩阵求逆运算，适用xilinx v6板卡(Inverse operation based on fpga matrix)

说明：  测绘技能大赛试题03 帮助需要进行比赛的同学解除困难(Question 03 of Surveying and Mapping Skills Competition)

基于C语言的MPICH程序，实现矩阵相乘(C-based MPICH program, to achieve matrix multiplication)

MPI.net tool for MPI setup for all applications that using MPI

CUDA平行的優化，以矩陣乘法為基礎。透過ShareMemory做加速(CUDA ShareMem)

应用背景  Apriori算法是一种最有影响的挖掘布尔关联规则频繁项集的算法。其核心是基于两阶段频集思想的递推算法。该关联规则在分类上属于单维、单层、布尔关联规则。在这里，所有支持度大于最小支持度的项集称为频繁项集，简称频集。  关键技术频繁项集挖掘采用spark实现的并行化挖掘算法。并行的apriori算法采用scala编写，并行的apriori采用java编写。里面会有较多函数使编程的内容。 

提出了一种以目标函数变化量作为评价函数的改进禁忌搜索算法,并进行了理论分析,然后将其与有效禁忌搜索算法作了性能比较。通过比较三个公共测试数据集的计算结果,验证了本文提出的禁忌搜索算法的可行性和有效性。(Proposed a change in the objective function as the evaluation function of the improved tabu search algorithm, and has carried on the theoretical analysis, then made a performance comparison with tabu search algorithm is effective. By comparing the calculation results of three test data set, and verified the feasibility and effectiveness of the proposed tabu search algorithm.)

CUDA是流行的并行计算的硬件和软件架构， 其高性能的可以应用到实际软件中的工具，包含一些学习CUDA的基本的一些资料，并且有本人的对别人讲过的PPT 已经一些总结(CUDA is a popular parallel computing hardware and software architecture, which can be applied to the actual high-performance software tools, including some of the study some of the basic information on CUDA, and there I talked to others in the PPT have been a number of summary)

LU算法的MPI串行实现，矩阵维数无要求，可在曙光4000A并行机运行(The LU algorithm MPI serial implementation matrix dimension requirements Dawning 4000A parallel machine running)