本文選題：Kubernetes + 資源調(diào)度�。� 參考：《電子科技大學(xué)》2017年碩士論文

【摘要】：以Docker為代表的容器技術(shù)在應(yīng)用的開發(fā)、發(fā)布和部署上具有便捷性和實(shí)用性,從出現(xiàn)之初便受到了業(yè)界的廣泛關(guān)注,由于Docker本身只注重于提供容器和鏡像,因此需要一個(gè)集成的容器云管理平臺(tái)高效地完成容器的編排部署、資源調(diào)度、服務(wù)發(fā)現(xiàn)、健康監(jiān)控等任務(wù)。Kubernetes憑借其強(qiáng)大的容器編排能力和輕量開源的特點(diǎn)成為了眾多容器集群調(diào)度系統(tǒng)的領(lǐng)跑者,然而Kubernetes的資源調(diào)度策略和系統(tǒng)自帶的調(diào)度算法都較為單一,在復(fù)雜的應(yīng)用場(chǎng)景下往往力不從心。本文在深入研究Kubernetes的核心技術(shù)后,對(duì)資源調(diào)度模塊進(jìn)行改進(jìn)和設(shè)計(jì),主要內(nèi)容如下:1.改進(jìn)了Kubernetes的資源模型。Kubernetes在進(jìn)行資源調(diào)度時(shí)只考量了CPU和內(nèi)存的影響,但是Pod若要正常運(yùn)行還需要進(jìn)行鏡像下載,與持久化存儲(chǔ)系統(tǒng)進(jìn)行數(shù)據(jù)交互,本文在原有模型上增加了鏡像下載速度和數(shù)據(jù)傳輸速度作為資源調(diào)度的考量因素。2.改進(jìn)了Kubernetes用戶綁定策略。Kubernetes的用戶綁定策略較為簡(jiǎn)單,本文在此基礎(chǔ)上設(shè)計(jì)了一種弱綁定策略,增加了用戶綁定的匹配規(guī)則,支持對(duì)表達(dá)式的匹配性周期檢測(cè)。3.設(shè)計(jì)搶占式調(diào)度策略。根據(jù)重啟策略將Pod劃分為三個(gè)優(yōu)先級(jí),當(dāng)宿主機(jī)資源不足時(shí),高優(yōu)先級(jí)Pod可以搶占低優(yōu)先級(jí)Pod的資源,有效的提高了高優(yōu)先級(jí)Pod的運(yùn)行比例。4.設(shè)計(jì)動(dòng)態(tài)負(fù)載均衡調(diào)度策略。Kubernetes默認(rèn)調(diào)度策略中Pod一經(jīng)調(diào)度便無法遷移,本文鑒于此設(shè)計(jì)了一種基于Kubernetes的動(dòng)態(tài)負(fù)載均衡改進(jìn)算法,該算法的靜態(tài)調(diào)度負(fù)責(zé)將待調(diào)度Pod隊(duì)列的每一個(gè)Pod調(diào)度到最符合其資源描述文件的節(jié)點(diǎn)上;動(dòng)態(tài)調(diào)度則通過監(jiān)控器定期將宿主機(jī)和Pod的運(yùn)行狀況反饋到調(diào)度器,調(diào)度器根據(jù)系統(tǒng)整體負(fù)載情況作出動(dòng)態(tài)調(diào)整,將負(fù)載較高的節(jié)點(diǎn)的一些Pod遷移到負(fù)載較低的節(jié)點(diǎn)上,以維持系統(tǒng)的整體負(fù)載均衡。
[Abstract]:The container technology represented by Docker has the convenience and practicability in the development, release and deployment of the application. From the beginning of it, the container technology has received extensive attention from the industry. Because Docker itself only pays attention to providing containers and mirrors, it needs an integrated container cloud management platform to efficiently complete the arrangement and deployment of the container, resource scheduling, Service discovery, health monitoring and other tasks.Kubernetes have become the leader of a large number of container cluster scheduling systems with its powerful container arrangement ability and light source and open source characteristics. However, the resource scheduling strategy of Kubernetes and the scheduling algorithm brought by the system are relatively simple. After studying the core technology of Kubernetes, the resource scheduling module is improved and designed. The main contents are as follows: 1. improved the resource model.Kubernetes of Kubernetes to consider only the influence of CPU and memory during the resource scheduling, but if Pod wants to run normal, it also needs to download the mirror image and make the data intersection with the persistent storage system. In the original model, this paper adds the speed of the image download and the data transmission speed as the consideration factor of the resource scheduling..2. improves the user binding strategy of the Kubernetes user binding strategy.Kubernetes. On this basis, this paper designs a weak binding strategy, adding the matching rules of the user binding, and supporting the expression. .3. design preemption scheduling strategy. According to reboot strategy, the Pod is divided into three priorities. When the host resource is insufficient, high priority Pod can seize the resources of low priority Pod, effectively improve the high priority Pod operation ratio.4. design dynamic load balancing scheduling strategy.Kubernetes default scheduling. In the strategy, Pod can not be migrated once it is dispatched. In this paper, a dynamic load balancing improvement algorithm based on Kubernetes is designed. The static scheduling of the algorithm is responsible for scheduling every Pod of the scheduled Pod queue to the node that is most consistent with its resource description file; the dynamic adjustment will regularly transport the host and Pod through the monitor. The line status is fed back to the scheduler. The scheduler makes dynamic adjustments based on the overall load condition of the system, and migrates some of the Pod with higher load to the lower load nodes to maintain the overall load balance of the system.
【學(xué)位授予單位】：電子科技大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP393.09

【參考文獻(xiàn)】

相關(guān)期刊論文 前9條

1 Zhenhua Li;Yun Zhang;Yunhao Liu;;Towards a Full-Stack Dev Ops Environment (Platform-as-a-Service) for Cloud-Hosted Applications[J];Tsinghua Science and Technology;2017年01期

2 毛祺;盧勝林;;基于Docker Swarm集群的容器遷移策略的實(shí)現(xiàn)[J];信息技術(shù);2016年09期

3 盧勝林;倪明;張翰博;;基于Docker Swarm集群的調(diào)度策略優(yōu)化[J];信息技術(shù);2016年07期

4 劉思堯;李強(qiáng);李斌;;基于Docker技術(shù)的容器隔離性研究[J];軟件;2015年04期

5 張建;謝天鈞;;基于Docker的平臺(tái)即服務(wù)架構(gòu)研究[J];信息技術(shù)與信息化;2014年10期

6 薛玉;;云計(jì)算環(huán)境下的資源調(diào)度優(yōu)化模型研究[J];計(jì)算機(jī)仿真;2013年05期

7 鄭海榮;高敬;;云計(jì)算技術(shù)與應(yīng)用[J];通信技術(shù);2013年04期

8 朱澤民;張青;;基于多維QoS和云計(jì)算的資源負(fù)載均衡調(diào)度研究[J];計(jì)算機(jī)測(cè)量與控制;2013年01期

9 林偉偉;齊德昱;;云計(jì)算資源調(diào)度研究綜述[J];計(jì)算機(jī)科學(xué);2012年10期

相關(guān)博士學(xué)位論文 前2條

1 劉琨;云計(jì)算負(fù)載均衡策略的研究[D];吉林大學(xué);2016年

2 王智明;云數(shù)據(jù)中心資源調(diào)度機(jī)制研究[D];北京郵電大學(xué);2012年

相關(guān)碩士學(xué)位論文 前8條

1 杜軍;基于Kubernetes的云端資源調(diào)度器改進(jìn)[D];浙江大學(xué);2016年

2 劉敏獻(xiàn);基于Docker的服務(wù)調(diào)用拓?fù)浞治龊托阅鼙O(jiān)控系統(tǒng)的設(shè)計(jì)與實(shí)現(xiàn)[D];浙江大學(xué);2016年

3 陳星宇;基于容器云平臺(tái)的網(wǎng)絡(luò)資源管理與配置系統(tǒng)設(shè)計(jì)與實(shí)現(xiàn)[D];浙江大學(xué);2016年

4 仇臣;Docker容器的性能監(jiān)控和日志服務(wù)的設(shè)計(jì)與實(shí)現(xiàn)[D];浙江大學(xué);2016年

5 孫庚澤;基于Docker的混合云應(yīng)用編排方案研究[D];西安電子科技大學(xué);2015年

6 梁俊杰;基于應(yīng)用容器的云資源調(diào)度研究與實(shí)現(xiàn)[D];電子科技大學(xué);2015年

7 余浩維;PaaS云中Web容器及調(diào)度的設(shè)計(jì)與實(shí)現(xiàn)[D];北京郵電大學(xué);2015年

8 張麗梅;基于負(fù)載均衡的云資源調(diào)度策略研究[D];寧夏大學(xué);2014年

，

本文編號(hào)：2039932