本文選題：MPI　切入點(diǎn)：OpenMP　出處：《成都理工大學(xué)》2012年碩士論文　論文類型：學(xué)位論文

【摘要】：隨著計(jì)算機(jī)應(yīng)用的范圍越來越廣，處理問題的規(guī)模越來越大，計(jì)算機(jī)硬件得到了迅速發(fā)展，近年來已經(jīng)進(jìn)入到多核體系結(jié)構(gòu)、個(gè)人高性能計(jì)算機(jī)、千萬億次并行機(jī)的發(fā)展階段。為了適應(yīng)迅速發(fā)展的計(jì)算機(jī)硬件和滿足各應(yīng)用領(lǐng)域?qū)τ?jì)算能力日益劇增的需求，并行計(jì)算技術(shù)得到了快速發(fā)展并廣泛地應(yīng)用于天體物理、流體力學(xué)、油藏建模、海洋環(huán)流、中長期天氣預(yù)報(bào)、地震數(shù)據(jù)處理、生物信息處理、計(jì)算機(jī)輔助設(shè)計(jì)、數(shù)據(jù)庫管理、圖像處理以及商用搜索引擎等領(lǐng)域。 根據(jù)處理器架構(gòu)和存儲器類型的不同，并行計(jì)算分為分布式并行計(jì)算和共享存儲并行計(jì)算。分布式并行計(jì)算一般采用消息傳遞模型，MPI（消息傳遞接口）是當(dāng)前的消息傳遞編程標(biāo)準(zhǔn)，具有可移植性好，高效率等優(yōu)點(diǎn)。而OpenMP是共享存儲模型的標(biāo)準(zhǔn)，對于描述單個(gè)SMP節(jié)點(diǎn)內(nèi)部處理器之間的通信更加有效。目前結(jié)合兩者優(yōu)勢的MPI+OpenMP多層次并行編程模型已成為主流編程技術(shù)，這種并行編程模型在多核體系結(jié)構(gòu)中能夠充分地發(fā)揮MPI與OpenMP之間的互補(bǔ)優(yōu)勢。 本文以合成孔徑雷達(dá)干涉測量（InSAR）技術(shù)中的基于網(wǎng)絡(luò)模型的統(tǒng)計(jì)費(fèi)用流相位解纏算法為應(yīng)用基礎(chǔ)，針對海量SAR數(shù)據(jù)處理面臨計(jì)算資源不足和相對運(yùn)算效率不高的問題，將統(tǒng)計(jì)費(fèi)用流相位解纏算法進(jìn)行并行化以提高相位解纏的運(yùn)算效率，為InSAR數(shù)據(jù)處理中其他環(huán)節(jié)的并行處理奠定基礎(chǔ)。本文首先研究了近幾十年來國內(nèi)外的并行計(jì)算技術(shù)以及一些經(jīng)典的并行編程模型，然后選擇MPI和OpenMP作為主要研究對象，分析它們的編程模型以及MPI+OpenMP多層次并行編程模型。在此基礎(chǔ)上，分析研究了統(tǒng)計(jì)費(fèi)用流相位解纏串行算法，挖掘該算法的內(nèi)在并行性，設(shè)計(jì)與實(shí)現(xiàn)了基于MPI的統(tǒng)計(jì)費(fèi)用流相位解纏并行算法，并進(jìn)一步分析了統(tǒng)計(jì)費(fèi)用流相位解纏串行算法中可細(xì)粒度并行計(jì)算的代碼，實(shí)現(xiàn)了基于MPI+OpenMP多層次并行編程的統(tǒng)計(jì)費(fèi)用流相位解纏并行算法。在研究過程中，分別對MPI環(huán)境和多層次并行編程MPI+OpenMP環(huán)境下的統(tǒng)計(jì)費(fèi)用流相位解纏并行算法進(jìn)行了不同計(jì)算規(guī)模的性能比較實(shí)驗(yàn)，驗(yàn)證了多層次并行編程具有良好的并行計(jì)算加速性能。 在實(shí)現(xiàn)基于MPI的統(tǒng)計(jì)費(fèi)用流相位解纏并行算法過程中，本文還研究了MPI-2中的并行I/O技術(shù)，根據(jù)相位解纏過程中I/O操作的特性，采用顯示偏移量的并行I/O方式進(jìn)行處理，，并通過實(shí)驗(yàn)驗(yàn)證了加入并行I/O技術(shù)的統(tǒng)計(jì)費(fèi)用流相位解纏并行算法的性能更好。 通過實(shí)驗(yàn)與分析表明，本文設(shè)計(jì)與實(shí)現(xiàn)的基于純MPI和基于MPI+OpenMP多層次并行的兩種統(tǒng)計(jì)費(fèi)用流相位解纏并行算法均有效地增加了相位解纏的效率，減少了解纏過程中內(nèi)存空間的開銷，并且在一定條件下，相同計(jì)算規(guī)模的MPI+OpenMP多層次并行算法性能優(yōu)于純MPI的并行算法。
[Abstract]:With the wide application of computer, the scale of dealing with the problem is becoming larger and larger, the computer hardware has been rapidly developed, in recent years has entered the multi-core architecture, personal high-performance computer, In order to adapt to the rapid development of computer hardware and meet the increasing demand for computing power in various application fields, parallel computing technology has been rapidly developed and widely used in astrophysics. Fluid mechanics, reservoir modeling, ocean circulation, medium and long term weather forecasting, seismic data processing, biological information processing, computer aided design, database management, image processing and commercial search engines. Depending on the processor architecture and memory type, Parallel computing is divided into distributed parallel computing and shared storage parallel computing. Generally, distributed parallel computing adopts message passing model (MPI), which is the current message passing programming standard, and has good portability. OpenMP is the standard of shared storage model, which is more effective in describing the communication between the internal processors of a single SMP node. At present, the multilevel parallel programming model of MPI OpenMP, which combines the advantages of both, has become the mainstream programming technology. This parallel programming model can take full advantage of the complementary advantages between MPI and OpenMP in multi-core architecture. Based on the statistical cost flow phase unwrapping algorithm based on network model in synthetic Aperture Radar Interferometry (SAR) technology, this paper aims at the problems of insufficient computing resources and low relative computational efficiency for massive SAR data processing. The statistical cost flow phase unwrapping algorithm is parallelized to improve the efficiency of phase unwrapping. This paper first studies the parallel computing technology and some classical parallel programming models at home and abroad in recent decades, and then chooses MPI and OpenMP as the main research objects. Their programming models and MPI OpenMP multilevel parallel programming models are analyzed. Based on this, the statistical cost flow phase unwrapping serial algorithm is analyzed and the inherent parallelism of the algorithm is explored. A phase unwrapping parallel algorithm based on MPI is designed and implemented, and the code of fine-grained parallel computation in the phase unwrapping serial algorithm of statistical cost flow is analyzed. The phase unwrapping parallel algorithm of statistical cost flow based on MPI OpenMP multilevel parallel programming is implemented. The performance comparison experiments of statistical cost flow phase unwrapping parallel algorithms in MPI environment and multi-level parallel programming MPI OpenMP environment are carried out, and the results show that multi-level parallel programming has good parallel computing acceleration performance. In the process of implementing the phase unwrapping parallel algorithm based on MPI, the parallel I / O technique in MPI-2 is also studied. According to the characteristics of I / O operation during phase unwrapping, the parallel I / O method is used to display the offset. The performance of phase unwrapping parallel algorithm with parallel I / O technique is proved to be better by experiments. The experiments and analysis show that the two parallel algorithms of phase unwrapping based on pure MPI and MPI OpenMP, which are based on pure MPI and MPI OpenMP, both increase the efficiency of phase unwrapping effectively. Under certain conditions, the performance of MPI OpenMP multi-level parallel algorithm with the same computational size is superior to that of pure MPI parallel algorithm.
【學(xué)位授予單位】：成都理工大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2012
【分類號】：TP338.6;TN957.52

【參考文獻(xiàn)】

相關(guān)期刊論文 前7條

1 盧麗君,廖明生,張路;分布式并行計(jì)算技術(shù)在遙感數(shù)據(jù)處理中的應(yīng)用[J];測繪信息與工程;2005年03期

2 陳華;范宜仁;鄧少貴;李智強(qiáng);;基于多核微機(jī)的微粒群并行算法[J];計(jì)算機(jī)工程與應(yīng)用;2010年13期

3 陳國良;孫廣中;徐云;龍柏;;并行計(jì)算的一體化研究現(xiàn)狀與發(fā)展趨勢[J];科學(xué)通報(bào);2009年08期

4 董麗麗;劉明生;袁香菊;;多核并行編程技術(shù)在中文分詞程序優(yōu)化中的應(yīng)用[J];計(jì)算機(jī)工程與設(shè)計(jì);2010年24期

5 張揚(yáng)奇;肖學(xué)福;高兵權(quán);劉金彪;;并行機(jī)中的Cache一致性技術(shù)研究[J];數(shù)字技術(shù)與應(yīng)用;2010年03期

6 王麗娟;蔣厚軍;達(dá)漢橋;廖明生;;統(tǒng)計(jì)費(fèi)用網(wǎng)絡(luò)流相位解纏并行處理[J];武漢大學(xué)學(xué)報(bào)(信息科學(xué)版);2010年09期

7 劉軼;張昕;李鶴;錢德沛;;多核處理器大規(guī)模并行系統(tǒng)中的任務(wù)分配問題及算法[J];小型微型計(jì)算機(jī)系統(tǒng);2008年05期

相關(guān)博士學(xué)位論文 前3條

1 陳永健;OpenMP編譯與優(yōu)化技術(shù)研究[D];清華大學(xué);2004年

2 章隆兵;提高機(jī)群OpenMP系統(tǒng)性能的關(guān)鍵技術(shù)研究[D];中國科學(xué)院研究生院（計(jì)算技術(shù)研究所）;2004年

3 黃春;面向分布共享存儲體系結(jié)構(gòu)的高效能OpenMP關(guān)鍵技術(shù)研究[D];國防科學(xué)技術(shù)大學(xué);2007年

相關(guān)碩士學(xué)位論文 前10條

1 游佐勇;OpenMP并行編程模型與性能優(yōu)化方法的研究及應(yīng)用[D];成都理工大學(xué);2011年

2 殷順昌;OpenMP并行程序性能分析[D];國防科學(xué)技術(shù)大學(xué);2006年

3 胡黎輝;金屬凝固過程溫度場模擬并行軟件開發(fā)[D];重慶大學(xué);2007年

4 孫守航;異構(gòu)多核處理器OpenMP編譯實(shí)現(xiàn)與優(yōu)化[D];國防科學(xué)技術(shù)大學(xué);2007年

5 汪鵬飛;并行蟻群算法及其應(yīng)用研究[D];西南交通大學(xué);2008年

6 楊尚琴;多層次并行算法與MPI-2新特性的研究及應(yīng)用[D];成都理工大學(xué);2009年

7 劉勝飛;OpenMP循環(huán)調(diào)度算法及SpMV多核并行化研究[D];中國科學(xué)院研究生院（軟件研究所）;2009年

8 陳秋麗;安全算法并行計(jì)算研究與實(shí)現(xiàn)[D];西安電子科技大學(xué);2010年

9 唐玲;OpenMP多線程負(fù)載均衡分析方法及調(diào)度策略研究[D];湖南大學(xué);2010年

10 陳林;基于多核架構(gòu)體系的圖像配準(zhǔn)并行計(jì)算方法研究[D];云南大學(xué);2010年

本文編號：1575582