本文選題：復(fù)雜網(wǎng)絡(luò)　切入點(diǎn)：軟件執(zhí)行網(wǎng)絡(luò)　出處：《燕山大學(xué)》2015年碩士論文　論文類型：學(xué)位論文

【摘要】：對(duì)一個(gè)實(shí)時(shí)的軟件系統(tǒng)來(lái)說(shuō),軟件中的一些關(guān)鍵實(shí)體在軟件運(yùn)行過(guò)程中起著重要作用,關(guān)乎軟件系統(tǒng)的質(zhì)量。軟件節(jié)點(diǎn)作為軟件實(shí)體關(guān)鍵的一部分,對(duì)它的重要性評(píng)估是軟件理解過(guò)程中一個(gè)必不可少的輔助手段。由于目前的重要性評(píng)估方法在不同環(huán)境下適用程度存在差異,適合軟件系統(tǒng)的重要節(jié)點(diǎn)發(fā)現(xiàn)方法和評(píng)估方法還不完善。為此,本文從復(fù)雜網(wǎng)絡(luò)角度,基于級(jí)聯(lián)失效的方法對(duì)軟件執(zhí)行網(wǎng)絡(luò)重要節(jié)點(diǎn)進(jìn)行了研究,主要內(nèi)容涉及如下:首先,由于軟件系統(tǒng)的靜態(tài)結(jié)構(gòu)難以揭示軟件系統(tǒng)全部特性,本文通過(guò)追蹤軟件執(zhí)行過(guò)程中的堆棧變化,以獲取軟件動(dòng)態(tài)執(zhí)行過(guò)程信息,對(duì)軟件動(dòng)態(tài)執(zhí)行過(guò)程進(jìn)行復(fù)雜網(wǎng)絡(luò)建模。其次,由于軟件執(zhí)行網(wǎng)絡(luò)中,一個(gè)或者少數(shù)幾個(gè)節(jié)點(diǎn)發(fā)生故障,會(huì)通過(guò)節(jié)點(diǎn)之間的耦合關(guān)系引起其它節(jié)點(diǎn)發(fā)生故障產(chǎn)生連鎖效應(yīng),最終將導(dǎo)致整個(gè)網(wǎng)絡(luò)的崩潰,帶來(lái)軟件安全問(wèn)題。本文在軟件執(zhí)行有向加權(quán)網(wǎng)絡(luò)模型基礎(chǔ)上,定義了節(jié)點(diǎn)的故障傳播能力及故障傳播率,并設(shè)計(jì)了故障的傳播策略,提出了一種級(jí)聯(lián)故障擴(kuò)散影響分析CFDA(Cascading Failure Diffusion Analyzing)算法來(lái)評(píng)估軟件在運(yùn)行時(shí)節(jié)點(diǎn)故障的影響。再次,本文基于級(jí)聯(lián)失效的方法,分別通過(guò)蓄意攻擊軟件執(zhí)行網(wǎng)絡(luò)中入點(diǎn)強(qiáng)度高的節(jié)點(diǎn)和傳播能力強(qiáng)的節(jié)點(diǎn),對(duì)比分析兩種攻擊策略產(chǎn)生級(jí)聯(lián)失效時(shí)節(jié)點(diǎn)的故障傳播范圍和受感染節(jié)點(diǎn)數(shù)量,以此來(lái)分析哪些節(jié)點(diǎn)在軟件中產(chǎn)生級(jí)聯(lián)失效影響更大,更應(yīng)受到保護(hù)。最后,本文分析了在節(jié)點(diǎn)遭受攻擊而引發(fā)級(jí)聯(lián)失效的情況下,不同節(jié)點(diǎn)故障對(duì)軟件執(zhí)行網(wǎng)絡(luò)的影響,以此評(píng)估節(jié)點(diǎn)的重要性,并給出了相應(yīng)的實(shí)驗(yàn)過(guò)程,驗(yàn)證了方法的可行性和適用性。
[Abstract]:In a real-time software system, some key entities in the software plays an important role in the software process, for the quality of software system. The software of nodes as part of a software entity key, to evaluate the importance of it is an indispensable means for software understanding process. Due to the importance evaluation currently there are differences in different environment suitability for important node software system discovery method and evaluation method is not perfect. Therefore, this paper from the perspective of complex network, the method based on cascading failure of software with an important node in network is studied, the main contents are as follows: firstly, because of all the involved properties of the software system is difficult to to reveal the static structure of the software system, the stack through the tracking software in the implementation process, to obtain information on the implementation of dynamic software, software dynamic. The process of complex network modeling. Secondly, the software implementation of the network, one or a few node failure, through the coupling between the nodes caused by other node failure have a ripple effect, will eventually lead to the collapse of the whole network, bring software security issues in this article. The software implementation based weighted network model on the definition of fault propagation and fault propagation rate of the node, and designed the fault propagation strategy, proposes a cascading failure diffusion effect analysis of CFDA (Cascading Failure Diffusion Analyzing) algorithm is used to evaluate the effect of software at runtime node failure. Again, this method based on cascading failure, respectively, through the node attack software implementation of network nodes and communication ability in high strength strong, comparative analysis of two kinds of attack strategies generated by cascading failure when the node Avoidance of transmission range and infected node number, in order to analyze the effect of cascading failure more which nodes are generated in the software, it should be protected. Finally, this paper analyzes the attacks of nodes caused by cascading failure conditions, effects of different node fault on software implementation of the network, in order to assess the importance of nodes, and gives the the corresponding experiment, verified the feasibility and applicability of the method.

【學(xué)位授予單位】：燕山大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2015
【分類號(hào)】：TP311.53;O157.5

【參考文獻(xiàn)】

相關(guān)期刊論文 前1條

1 馬于濤;何克清;李兵;劉婧;;網(wǎng)絡(luò)化軟件的復(fù)雜網(wǎng)絡(luò)特性實(shí)證[J];軟件學(xué)報(bào);2011年03期

相關(guān)碩士學(xué)位論文 前1條

1 蘭文輝;基于復(fù)雜網(wǎng)絡(luò)的可信軟件研究[D];大連理工大學(xué);2010年

，

本文編號(hào)：1568409