隨著雲端科技不斷的進步和普及，我們對雲端服務的需求也越來越多，因此市面上也出現越來越多的雲端平台提供雲端服務，例如:GCP(Google cloud platform)、Microsoft Azure、Amazon AWS、IBM bluemix、OpenStack…..等等。

雖然雲端環境可以簡單的建立，但是資源的分配和資源最佳化的使用依然還是一個具有挑戰性的議題，例如:虛擬機器應該要基於詳細的硬體需求去部屬為目標，換句話說，系統必須針對不同型態的虛擬機器去配置在最符合此虛擬機的實體機中去執行。再來是，如果當系統認為實體機已經沒有足夠的資源能供給配置虛擬機時，我們可以透過搬移虛擬機到其他有足夠資源的實體機中，讓原本不夠資源配置的實體機多出了空間可以配置其他的虛擬機，讓有限的資源中做更有效的利用，而在OpenStack中只提供了基本對資源分配以及搬移虛擬機例子的指令，除此之外，OpenStack的kilo版本的排程方法只採用了記憶體空間作為考量，和沒有一個針對搬移虛擬機實例的方法，所以我們還有空間可以改善OpenStack對分配虛擬機的排程和搬移虛擬機實例的方法，這也是這篇論文主要討論的重點。

在這篇論文一開始我們會先介紹在OpenStack NOVA運算模組裡的nova-schedule和搬移虛擬機，nova-schedule的主要功能是用於分配硬體資源給虛擬機，而搬移虛擬機是可以將虛擬機搬到另一個實體機中。在OpenStack使用者可以根據dashboard module輸入參數去客制化虛擬機的規格，再來我們利用[2]中的Physical host Aggregate Filter和resource usage aware scheduler去分配虛擬機的資源到實體機，之後我們提出了虛擬機搬移之OpenStack排程策略方法讓我們可以透過搬移虛擬機而不會過載的情況下配置更多的虛擬機，使資源的分配更有效率。

As the demands for cloud services increase, the cloud technology advances and popularizes, there are more and more cloud platforms in the market (which can deploy the cloud environment with low-cost) such as: GCP(Google cloud platform)、 Microsoft Azure、Amazon AWS、IBM bluemix、OpenStack…etc. 
Though cloud environment can be established easily, resource distribution and optimization are still challenging topics. For instance, if a virtual machine needs to be deployed with specific hardware requirements, then the system needs to configure and run it on the most ideal physical host according to the capability of the physical hosts. In addition, if the physical host is out of resource to serve the specific virtual machine, the system can move the virtual machine to other physical hosts, so the original physical host can serve other virtual machines. By virtual machine migration, the system can maximize resource usage efficiently under the limited resources environment. OpenStack provides only the basic commands to distribute and move virtual machine, and only memory space is considered in the Kilo release of OpenStack while schedule the virtual machine, no specific strategy for virtual machine migration. This means there is still room to improvising OpenStack when it comes to distributing the schedule for virtual machines as well as instantiations of moving virtual machines, which is also the focus of this discussion.
Therefore, this paper will first review the nova schedule and VM migration in OpenStack; the main function of nova-schedule is to determine how to dispatch computer requests, while migrate VM instance can move VM instances to other physical hosts. OpenStack users can then input parameters according to the dashboard module to customize specifications of the virtual machine. We use the Physical host Aggregate Filter and resource usage aware scheduler to assign virtual machine to the physical host. After that, we propose a method to migrate virtual machines between physical hosts while guarantee no physical host is overloading, thus improving the efficiency of resource allocation.

目錄
第一章	緒論	1
1.1	前言	1
1.2	動機與目的	2
1.3	論文章節架構	4
第二章	背景知識與相關文獻	5
2.1	雲端運算	5
2.1.1	定義	6
2.1.2	三種服務模式	6
2.1.3	虛擬化技術(Virtualization)	7
2.2	OpenStack	8
2.2.1	起源與發展	8
2.2.2	核心套件與系統架構	9
2.2.3	Filter-Weight排程演算法	12
(1)	第一階段-Filter	13
(2)	第二階段-Weight	14
2.3	資源負載感知之排程機制	15
2.3.1	Filter	16
2.3.1.1	Host Aggregate Filter	16
2.3.2	Weight	18
2.3.2.1	CPU Weigher	18
2.3.2.2	Disk Weigher	20
2.3.2.3	Bandwidth Weigher	22
2.3.2.4	Min-Max 正規化	24
第三章	虛擬機搬移之排程策略	25
3.1	第一階段-目標來源主機篩選	27
3.1.1	CPU Weigher	28
3.1.2	Memory Weigher	29
3.1.3	Disk Weigher	30
3.1.4	Bandwidth Weigher	32
3.2	第二階段-目標虛擬機篩選	34
3.2.1	CPU Weigher:	35
3.2.2	Memory Weigher:	36
3.2.3	Disk Weigher	37
3.2.4	Bandwidth Weigher	38
3.3	第三階段-目的主機篩選	39
3.3.1	CPU Weigher:	40
3.3.2	Memory Weigher:	41
3.3.3	Disk Weigher:	42
3.3.4	Bandwidth Weigher:	43
3.4	整體流程整合	44
第四章	實驗環境	50
4.1 實體環境	50
4.2 虛擬機性質與規格	51
4.3 方法驗證	53
第五章	模擬結果與效能分析	60
5.1	模擬內容	60
5.2	模擬環境測試	61
5.2.1	實驗Ⅰ-數據與分析	62
5.2.2	實驗Ⅱ-數據與分析	66
5.2.2.1模擬七種性質虛擬機	67
5.3	主要貢獻	69
第六章	結論與未來展望	70
參考文獻	71

圖目錄

圖2.1	雲端運算概觀[4]	5
圖2.2	各企業與研究機構對雲端運算之定義[5]	6
圖2.3	OpenStack核心系統架構[9]	10
圖2.4	OpenStack排程示意圖	13
圖2.5	Weight階段示意圖[11]	15
圖2.6	Host Aggregate示意圖	17
圖3.1 第一階段-目標來源主機篩選	45
圖3.2 	第二階段-目標虛擬機篩選	47
圖3.3 第三階段-目的主機篩選	49
圖5.1	實驗Ⅰ-第一種進入順序的結果	63
圖5.2	實驗Ⅰ-第二種進入順序的結果	64
圖5.3	實驗Ⅰ-第三種進入順序的結果	65

表目錄

表2.1	CPU負載權重公式參數	19
表2.2	硬碟負載權重公式參數	21
表2.3	網路負載權重公式參數	23
表2.4	Min-Max正規化公式參數	24
表3.1 CPU權重公式-目標來源主機篩選	28
表3.2 記憶體權重公式-目標來源主機篩選	29
表3.3 硬碟權重公式-目標來源主機篩選	31
表3.4 頻寬權重公式-目標來源主機篩選	33
表3.5 CPU權重公式-目標虛擬機篩選	35
表3.6 記憶體權重公式-目標虛擬機篩選	36
表3.7 硬碟使用率權重公式-目標虛擬機篩選	37
表3.8 硬碟使用率權重公式-目標虛擬機篩選	38
表3.9 CPU 權重公式-目的主機篩選	40
表3.10 記憶體權重公式-目的主機篩選	41
表3.11 硬碟使用率權重公式-目的主機篩選	42
表3.12 記憶體權重公式-目的主機篩選	43
表4.1 實體主機規格	50
表4.2	虛擬機規格	52
表4.3虛擬機的排程順序	54
表4.4. 資源負載感知排程機制資源使用情況	55
表4.5 部屬在實體機上的虛擬機	55
表4.6 第一階段正規化的總和	56
表4.7 第三階段的結果	57
表4.4. 資源負載感知排程機制資源使用情況	58
表4.8 搬移過後的實體機資源使用情況	58
表4.9 搬移過後再配置雲端儲存虛擬機的使用情況	59
表5.1	實驗Ⅱ-模擬環境主機規格	67
表5.2	模擬Ⅱ-造成過載主機的虛擬機數量	68

[1]	OpenStack (https://www.OpenStack.org/)
[2]	Yi-Ting Lai, Pei-wen Hsiao, Wei-Tsong Lee, HsinWen Wei, Vooi-Voon Yap,” 11 Resource Usage Aware Scheduling In OpenStack” ,IEEE Region on Ten Conference(TENCON) ,2017
[3]	黃重憲, "淺談雲端運算(Cloud Computing)", 國立台灣大學計算機及資訊網路中心電子報第0008期, 2009/3/20
[4]	雲端運算
(https://zh.wikipedia.org/wiki/%E9%9B%B2%E7%AB%AF%E9%81%8B%E7%AE%97)
[5]	Cloud Open Lab
(https://www.cloudopenlab.org.tw/ccipo_industryDefinition.do)
[6]	黃明華, "扮演雲端運算關鍵基礎 開源虛擬化技術概觀", 2014/12/10
(http://www.netadmin.com.tw/article_content.aspx?sn=1412020002)
[7]	OpenStack 雲端開源平台技術整理
(https://www.gitbook.com/book/kairen/OpenStack-liberty/details)
[8]	Wikipedia–OpenStack (https://zh.wikipedia.org/wiki/OpenStack)
[9]	OpenStack Architecture
(http://www.netadmin.com.tw/article_content.aspx?sn=1504070005&jump=2)
[10]	張景皓, "快速認識OpenStack技術架構", 2013/6/21
(http://www.ithome.com.tw/node/81095)
[11]	 OpenStack Scheduling (https://docs.OpenStack.org/kilo/config-  reference/content/section_compute-scheduler.html)
[12]	Yu Huanle, Shi Weifeng, Bai Tingting, "An OpenStack-based resource  optimization scheduling framework", Sixth International Symposium on Computational Intelligence and Design, 2013, pp. 261-264
[13]	Michael Scharf, Manuel Stein, Thomas Voith, Volker Hilt, "Network-aware Instance Scheduling in OpenStack", 2015 24th International Conference on Computer Communication and Networks (ICCCN), pp.1-6
[14]	Lianhao Lu, Yingxin Chen, "Utilization-based Scheduling in OpenStack* Compute (Nova) ", Intel Tech Tip, May 2017.
[15]	Shalmali Sahasrabudhe, Shilpa S. Sonawani, "Improved Filter-weight Algorithm for Utilization aware Resource Scheduling in OpenStack", 2015 International Conference on Information Processing (ICIP), Dec 16-19, pp. 43-47
[16]	Michael Rini Soosai, B.N.Jagdale, "Efficient Time Based Scheduler For Implementing Reservation Plan in Cloud", 2014 IEEE Global Conference on Wireless Computing and Networking (GCWCN), pp. 165-168.
[17]	Dan Tao, Zhaowen Lin, and Bingxu Wang, "Load Feedback-Based Resource Scheduling and Dynamic Migration-Based Data Locality for Virtual Hadoop Clusters in OpenStack-Based Clouds", 2017 20th Conference on Innovations in Clouds, Internet and Networks (ICIN), pp.149-159
[18]	Aditya Makwe1, Priyesh Kanungo2, "Scheduling in Cloud Computing Environment Using Analytic Hierarchy Process Model", IEEE International Conference on Computer, Communication and Control (IC4-2015), pp. 1-4.
[19]	Vatsal Gajera, Shubham, Rishabh Gupta, Prasanta K. Jana, "An effective Multi-Objective task scheduling algorithm using Min-Max normalization in cloud computing", 2016 2nd International Conference on Applied and Theoretical Computing and Communication Technology (iCATccT), pp. 812-816