【摘要】：近些年來(lái),隨著復(fù)雜網(wǎng)絡(luò)科學(xué)的不斷發(fā)展,學(xué)術(shù)界對(duì)于復(fù)雜網(wǎng)絡(luò)的研究逐漸深入,因?yàn)閺?fù)雜網(wǎng)絡(luò)的研究對(duì)很多領(lǐng)域的研究都具有指導(dǎo)意義,像其在社會(huì)學(xué)中展現(xiàn)出的價(jià)值或者對(duì)于傳播學(xué)的研究意義,其應(yīng)用也越來(lái)越廣泛。但是隨著社會(huì)進(jìn)步以及科技發(fā)展,網(wǎng)絡(luò)規(guī)模呈現(xiàn)出指數(shù)型增長(zhǎng),數(shù)據(jù)規(guī)模更加龐大,同時(shí)也面臨著數(shù)據(jù)處理緩慢甚至計(jì)算不出的問(wèn)題,如何更快更高效的處理這些數(shù)據(jù)成為了近期復(fù)雜網(wǎng)絡(luò)研究的重大挑戰(zhàn)之一。并行計(jì)算技術(shù)為復(fù)雜網(wǎng)絡(luò)的高效計(jì)算提供了可能。當(dāng)前的一些主流計(jì)算框架由于具有一定的限制,比如MapReduce在迭代次數(shù)多的情況下并沒(méi)有展現(xiàn)出很好的優(yōu)勢(shì),而復(fù)雜網(wǎng)絡(luò)的計(jì)算的特點(diǎn)之一就是具有較多的迭代次數(shù),因此在復(fù)雜網(wǎng)絡(luò)的計(jì)算方面并沒(méi)有顯示出很好的優(yōu)勢(shì)。志愿計(jì)算的核心思想就是將存在于網(wǎng)絡(luò)中的空閑資源利用,共同參與分布式計(jì)算。本文基于志愿計(jì)算的思想,以ICE中間件作為通信媒介,構(gòu)建了針對(duì)于大規(guī)模復(fù)雜網(wǎng)絡(luò)計(jì)算的松耦合分布式計(jì)算框架,用于對(duì)復(fù)雜網(wǎng)絡(luò)的分析和計(jì)算,并將其命名為DCBV框架�？蚣苤饕捎昧巳蝿�(wù)隊(duì)列的思想和方法,在Master/Worker模式的基礎(chǔ)上進(jìn)行設(shè)計(jì),增加中間層MiddleWare,將網(wǎng)絡(luò)內(nèi)的多臺(tái)空閑機(jī)器動(dòng)態(tài)的結(jié)合起來(lái)共同參與復(fù)雜網(wǎng)絡(luò)的分析和計(jì)算。除此之外,基于對(duì)平均最短路徑算法的改進(jìn),對(duì)設(shè)計(jì)的DCBV計(jì)算框架的原型實(shí)現(xiàn)進(jìn)行了實(shí)驗(yàn)和評(píng)估。實(shí)驗(yàn)結(jié)果表明,本文設(shè)計(jì)的框架可以更加高效的計(jì)算出復(fù)雜網(wǎng)絡(luò)的相關(guān)參數(shù),并且有很好的容錯(cuò)性,可以隨時(shí)調(diào)整計(jì)算節(jié)點(diǎn)的個(gè)數(shù),同時(shí)可以隨時(shí)動(dòng)態(tài)調(diào)整每個(gè)計(jì)算節(jié)點(diǎn)分配的線程數(shù)以及根據(jù)其分配的線程數(shù)為不同計(jì)算節(jié)點(diǎn)分配任務(wù)集。改進(jìn)的平均最短路徑算法相比于之前的算法更加高效,并將改進(jìn)后的算法在松耦合的計(jì)算網(wǎng)絡(luò)框架中實(shí)現(xiàn),獲得良好的加速比。
[Abstract]:In recent years, with the continuous development of complex network science, the academic research on complex network is gradually in-depth, because the study of complex network is of guiding significance to many fields of research. Such as its value in sociology or the significance of communication research, its application is becoming more and more extensive. However, with the progress of society and the development of science and technology, the scale of the network is increasing exponentially, and the scale of the data is even larger. At the same time, it is also facing problems that data processing is slow or even impossible to calculate. How to process these data faster and more efficiently has become one of the major challenges in recent complex network research. Parallel computing technology provides the possibility for efficient computing of complex networks. Some of the current mainstream computing frameworks have some limitations, such as MapReduce does not show a good advantage in the case of a large number of iterations, and one of the characteristics of complex network computing is that it has a large number of iterations. Therefore, the computation of complex network does not show a good advantage. The core idea of voluntary computing is to utilize the free resources existing in the network and participate in distributed computing together. Based on the idea of voluntary computing, this paper constructs a loosely coupled distributed computing framework for large-scale complex network computing using ICE middleware as a communication medium. It is used to analyze and calculate complex networks, and it is named DCBV framework. The framework mainly adopts the idea and method of task queue, designs on the basis of Master/Worker pattern, and adds the middle layer MiddleWare, to participate in the analysis and calculation of complex network together with the dynamic combination of many idle machines in the network. In addition, based on the improvement of the average shortest path algorithm, the prototype implementation of the designed DCBV computing framework is tested and evaluated. The experimental results show that the framework designed in this paper can calculate the parameters of the complex network more efficiently, and has good fault tolerance, and can adjust the number of nodes at any time. At the same time, the number of threads allocated by each computing node can be adjusted dynamically at any time, and the task set for different computing nodes can be assigned according to the number of threads allocated by each computing node. Compared with the previous algorithm, the improved average shortest path algorithm is more efficient, and the improved algorithm is implemented in the loosely coupled computing network framework, and a good speedup is obtained.
【學(xué)位授予單位】：北京化工大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2015
【分類(lèi)號(hào)】：O157.5

【參考文獻(xiàn)】

相關(guān)期刊論文 前5條

1 周濤;張子柯;陳關(guān)榮;汪小帆;史定華;狄增如;樊瑛;方錦清;韓筱璞;劉建國(guó);劉潤(rùn)然;劉宗華;陸君安;呂金虎;呂琳媛;榮智海;汪秉宏;許小可;章忠志;;復(fù)雜網(wǎng)絡(luò)研究的機(jī)遇與挑戰(zhàn)[J];電子科技大學(xué)學(xué)報(bào);2014年01期

2 吳長(zhǎng)茂;張聰品;張慧云;王娟;;CUDA平臺(tái)下多核GPU高性能并行編程研究[J];河南機(jī)電高等專(zhuān)科學(xué)校學(xué)報(bào);2011年01期

3 張俊軍;章旋;;ICE中間件技術(shù)及其應(yīng)用研究[J];計(jì)算機(jī)與現(xiàn)代化;2012年05期

4 許楨;;關(guān)于CPU+GPU異構(gòu)計(jì)算的研究與分析[J];科技信息;2010年17期

5 張保;曹海軍;董小社;李丹;胡雷鈞;;面向圖形處理器重疊通信與計(jì)算的數(shù)據(jù)劃分方法[J];西安交通大學(xué)學(xué)報(bào);2011年04期

，

本文編號(hào)：2456271