發(fā)布時(shí)間：2018-07-29 12:40

【摘要】：云計(jì)算技術(shù)已然成為當(dāng)今最熱門的網(wǎng)絡(luò)技術(shù)之一.云計(jì)算技術(shù)的興起,既是信息技術(shù)迅速發(fā)展的產(chǎn)物,也是人類社會(huì)對(duì)生活工作提出更高要求的體現(xiàn).云計(jì)算技術(shù)虛化了個(gè)人計(jì)算機(jī)的概念,而是通過第三方來實(shí)現(xiàn)計(jì)算機(jī)的存儲(chǔ)和計(jì)算任務(wù),然后通過按需付費(fèi)的方式提供給大眾使用.因此在第三方數(shù)據(jù)中心中如何快速高效的調(diào)度和使用巨大的資源,已經(jīng)成為云計(jì)算技術(shù)發(fā)展的關(guān)鍵.首先,本文將粒子群算法成功的應(yīng)用于云計(jì)算任務(wù)調(diào)度中,為了避免標(biāo)準(zhǔn)粒子群算法易陷入局部最優(yōu)的缺陷,因此引入了切比雪夫混沌擾動(dòng)策略,通過擾動(dòng)策略使得粒子群算法在運(yùn)算后期有能力跳出局部最優(yōu),使得粒子群算法可以得到更好的全局尋優(yōu)結(jié)果.通過運(yùn)用云計(jì)算仿真平臺(tái)Cloudsim進(jìn)行驗(yàn)證,實(shí)驗(yàn)結(jié)果表明改進(jìn)后的粒子群算法與其他一些傳統(tǒng)算法相比,在云計(jì)算任務(wù)調(diào)度中可以更短的時(shí)間內(nèi)獲得較好的調(diào)度結(jié)果.其次,本文在引入切比雪夫混沌擾動(dòng)策略的同時(shí),還加入了動(dòng)態(tài)慣性權(quán)重策略,使得改進(jìn)后的粒子群算法既有能力跳出局部最優(yōu),還可以根據(jù)實(shí)際問題動(dòng)態(tài)的調(diào)節(jié)自身全局搜索和局部搜素的能力.并將改進(jìn)后的算法應(yīng)用于云計(jì)算任務(wù)調(diào)度中,通過運(yùn)用云計(jì)算仿真平臺(tái)Cloudsim進(jìn)行驗(yàn)證,實(shí)驗(yàn)結(jié)果表明改進(jìn)后的算法比上述的改進(jìn)算法具有更優(yōu)異的調(diào)度結(jié)果,且所用的時(shí)間更短.最后,對(duì)多目標(biāo)粒子群算法進(jìn)行學(xué)習(xí)和研究,并應(yīng)用于云計(jì)算任務(wù)調(diào)度中.通過引入動(dòng)態(tài)慣性權(quán)重策略以及自適應(yīng)進(jìn)化學(xué)習(xí)策略,將多目標(biāo)粒子群算法進(jìn)行改進(jìn).通過運(yùn)用云計(jì)算仿真平臺(tái)Cloudsim進(jìn)行驗(yàn)證,實(shí)驗(yàn)結(jié)果表明改進(jìn)后的多目標(biāo)粒子群算法在多目標(biāo)云計(jì)算任務(wù)調(diào)度中在較短的時(shí)間內(nèi)可以獲得較好的調(diào)度結(jié)果.
[Abstract]:Cloud computing technology has become one of the most popular network technologies. The rise of cloud computing technology is not only the product of the rapid development of information technology, but also the embodiment of human society to put forward higher requirements for life and work. Cloud computing technology is a virtual concept of personal computers, but through a third party to achieve computer storage and computing tasks, and then through on-demand payment to the public to use. Therefore, how to quickly and efficiently schedule and use huge resources in third-party data centers has become the key to the development of cloud computing technology. First of all, particle swarm optimization algorithm is successfully applied to cloud computing task scheduling. In order to avoid the defect that standard particle swarm optimization algorithm is easy to fall into local optimum, Chebyshev chaos perturbation strategy is introduced. The PSO algorithm is able to jump out of the local optimum in the later stage of operation by perturbation strategy, so that the PSO algorithm can get better global optimization results. The experimental results show that the improved particle swarm optimization algorithm can obtain better scheduling results in a shorter time than other traditional algorithms by using cloud computing simulation platform Cloudsim. Secondly, the Chebyshev chaos perturbation strategy is introduced, and the dynamic inertial weight strategy is added, which makes the improved particle swarm optimization algorithm have the ability to jump out of the local optimum. The ability of global search and local search can be adjusted dynamically according to the actual problem. The improved algorithm is applied to the task scheduling of cloud computing and verified by the cloud computing simulation platform Cloudsim. The experimental results show that the improved algorithm has better scheduling results than the above improved algorithm and the time used is shorter. Finally, the multi-objective particle swarm optimization algorithm is studied and applied to cloud computing task scheduling. By introducing dynamic inertial weight strategy and adaptive evolutionary learning strategy, the multi-objective particle swarm optimization algorithm is improved. By using cloud computing simulation platform Cloudsim, the experimental results show that the improved multi-objective particle swarm optimization algorithm can obtain better scheduling results in a short time in multi-objective cloud computing task scheduling.
【學(xué)位授予單位】：北方民族大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP18

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 張佩云;鳳麒;;一種云計(jì)算環(huán)境下的工作流雙向調(diào)度方法[J];計(jì)算機(jī)科學(xué);2015年S2期

2 吳麗;余文春;;基于多服務(wù)器最優(yōu)配置的云計(jì)算利潤(rùn)最大化技術(shù)研究[J];計(jì)算機(jī)應(yīng)用研究;2015年01期

3 王建林;吳佳歡;張超然;趙利強(qiáng);于濤;;基于自適應(yīng)進(jìn)化學(xué)習(xí)的約束多目標(biāo)粒子群優(yōu)化算法[J];控制與決策;2014年10期

4 羅亮;吳文峻;張飛;;面向云計(jì)算數(shù)據(jù)中心的能耗建模方法[J];軟件學(xué)報(bào);2014年07期

5 劉川意;林杰;唐博;;面向云計(jì)算模式運(yùn)行環(huán)境可信性動(dòng)態(tài)驗(yàn)證機(jī)制[J];軟件學(xué)報(bào);2014年03期

6 任昱;李青榮;;基于VMware vSphere虛擬化資源管理平臺(tái)研究[J];計(jì)算機(jī)應(yīng)用與軟件;2012年05期

7 王晟;趙壁芳;;云計(jì)算中MapReduce技術(shù)研究[J];通信技術(shù);2011年12期

8 羅軍舟;金嘉暉;宋愛波;東方;;云計(jì)算:體系架構(gòu)與關(guān)鍵技術(shù)[J];通信學(xué)報(bào);2011年07期

9 華夏渝;鄭駿;胡文心;;基于云計(jì)算環(huán)境的蟻群優(yōu)化計(jì)算資源分配算法[J];華東師范大學(xué)學(xué)報(bào)(自然科學(xué)版);2010年01期

10 陳康;鄭緯民;;云計(jì)算:系統(tǒng)實(shí)例與研究現(xiàn)狀[J];軟件學(xué)報(bào);2009年05期

，

本文編號(hào)：2152699