本文選題：并行計(jì)算 + MPI��； 參考：《吉林大學(xué)》2014年碩士論文

【摘要】：科學(xué)技術(shù)發(fā)展至今，科學(xué)理論、科學(xué)計(jì)算、科學(xué)實(shí)驗(yàn)、全球氣侯預(yù)報(bào)、核聚變等粒子模型、大型計(jì)量經(jīng)濟(jì)模型、人類基因等，這些重要的科學(xué)技術(shù)課題中爆炸式增長(zhǎng)的數(shù)據(jù)計(jì)算量必然向傳統(tǒng)的Von Neuman串行計(jì)算技術(shù)提出挑戰(zhàn)，對(duì)高性能計(jì)算提出巨大的需求。 在如圖形圖像處理和分析、數(shù)字信號(hào)濾波計(jì)算、數(shù)值天氣預(yù)報(bào)、網(wǎng)絡(luò)路由選擇、智能交通系統(tǒng)最短路徑、線性規(guī)劃、傅里葉變換等眾多數(shù)值處理問題的理論模型中，矩陣是描述和存儲(chǔ)大規(guī)模數(shù)據(jù)最常用的邏輯結(jié)構(gòu)，相關(guān)應(yīng)用中，和矩陣緊密相連的求解問題不勝枚舉，因此，有關(guān)矩陣的運(yùn)算是許多科學(xué)和工程計(jì)算中最重要也是最基本的技術(shù)之一，近二十年來這方面的并行算法研究一直十分活躍。 本文運(yùn)用虛擬機(jī)技術(shù)、網(wǎng)絡(luò)技術(shù)、操作系統(tǒng)等有關(guān)理論構(gòu)建了一個(gè)由以太網(wǎng)互聯(lián)四臺(tái)虛擬機(jī)的機(jī)群系統(tǒng)，每個(gè)虛擬機(jī)的操作系統(tǒng)為L(zhǎng)inux的ubuntu版本，整個(gè)并行平臺(tái)為MPICH2并行環(huán)境。全文的并行算法研究工作都是以此平臺(tái)為基礎(chǔ)進(jìn)行的。 文中詳細(xì)介紹了并行編譯系統(tǒng)、并行算法的含義和結(jié)構(gòu)、MPI并行編程技術(shù)、并行性能評(píng)價(jià)指標(biāo)和并行基礎(chǔ)應(yīng)用。 本文基于抽象的計(jì)算模型，通過并行化串行算法，改進(jìn)已有并行算法，以及直接從問題本身出發(fā)研究新的算法這三種方式對(duì)矩陣乘的并行算法進(jìn)行研究，并以不同的數(shù)據(jù)劃分方式將算法分為帶狀劃分算法和棋盤劃分算法兩大類，詳細(xì)闡明了帶狀劃分算法中的行列、行行、列列、列行劃分算法和棋盤劃分算法中的Fox、Cannon算法。 在算法的實(shí)現(xiàn)過程中，本文著眼于對(duì)并行程序性能的不斷優(yōu)化。并行程序的優(yōu)化分為串行部分的優(yōu)化和并行部分的優(yōu)化兩方面。將并行程序在構(gòu)建的機(jī)群環(huán)境中進(jìn)行測(cè)試，分析和對(duì)比運(yùn)行效率。 本文構(gòu)建了一個(gè)完整的小型虛擬機(jī)群系統(tǒng)，并以其為平臺(tái)運(yùn)用并行技術(shù)解決了具體的應(yīng)用問題。 通過調(diào)用并行庫(kù)，，將矩陣乘算法的設(shè)計(jì)思路應(yīng)用到最短路徑的求解以及卷積運(yùn)算中去，設(shè)計(jì)并實(shí)現(xiàn)相應(yīng)的并行算法，取得了較好的效果。
[Abstract]:So far, scientific theories, scientific calculations, scientific experiments, global weather forecasting, nuclear fusion and other particle models, large-scale econometric models, human genes, etc. The explosive increase of data computation in these important scientific and technological subjects will inevitably challenge the traditional Von Neuman serial computing technology and bring forward a huge demand for high performance computing. In the theoretical models of many numerical processing problems, such as graphic image processing and analysis, digital signal filtering calculation, numerical weather forecast, network routing, intelligent transportation system shortest path, linear programming, Fourier transform, etc. Matrix is the most commonly used logical structure to describe and store large scale data. The computation of matrices is one of the most important and basic techniques in many scientific and engineering computations. The study of parallel algorithms in this field has been very active in the past two decades. This paper constructs a cluster system of four virtual machines interconnected by Ethernet using virtual machine technology, network technology and operating system theory. The operating system of each virtual machine is ubuntu version of Linux, and the whole parallel platform is MPICH2 parallel environment. The research work of parallel algorithms is based on this platform. The parallel compiling system, the meaning and structure of parallel algorithm, MPI parallel programming technology, parallel performance evaluation index and parallel basic application are introduced in detail in this paper. Based on the abstract computing model, this paper studies the parallel algorithm of matrix multiplication by parallelizing the serial algorithm, improving the existing parallel algorithm, and studying the new algorithm directly from the problem itself. The algorithms are divided into two categories: banded partition algorithm and chessboard partition algorithm by different data partitioning methods. The row, column, column row partition algorithm and the Fox-Cannon algorithm in the chessboard partitioning algorithm are expounded in detail. In the implementation of the algorithm, this paper focuses on the continuous optimization of the performance of parallel programs. The optimization of parallel program is divided into two aspects: the optimization of serial part and the optimization of parallel part. The parallel program is tested in the cluster environment, and the efficiency is analyzed and compared. In this paper, a complete small virtual machine swarm system is constructed, and the parallel technology is used to solve the specific application problem. By calling the parallel library, the design idea of matrix multiplication algorithm is applied to the solution of shortest path and convolution operation, and the corresponding parallel algorithm is designed and implemented, and good results are obtained.
【學(xué)位授予單位】：吉林大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2014
【分類號(hào)】：O151.21;TP302

【參考文獻(xiàn)】

相關(guān)期刊論文 前1條

1 張艷華;劉祥港;;一種基于MPI與OpenMP的矩陣乘法并行算法[J];計(jì)算機(jī)與現(xiàn)代化;2011年07期

本文編號(hào)：1914661