虛擬化技術在雲端計算中扮演舉足輕重的角色，它是雲端計算能夠實現的關鍵之一。近幾年來，容器虛擬化技術(Container-based virtualization)相當熱門，越來越多的公司在他們的產品中導入此技術 ; 相對於以往常使用的硬體虛擬化技術(Hardware virtualization)，容器虛擬化技術實現了更輕量以及更快速的虛擬方式，其中，最受歡迎使用此技術的方式是透過Docker將應用服務封裝進容器中執行，並用Kubernetes管理多台主機上所運行的Docker容器。由於不同的應用服務對於運算資源以及服務質量(Quality of Service，QoS)的需求也會不同，在總體資源有限的情況下，如何動態分配資源給這些運行中的應用服務成為了一個相當重要的議題，然而在原本的Kubernetes環境中，動態資源分配的機制並不是非常完善，因此，在本篇論文中，我們基於Kubernetes的環境，實作了一個完善的動態資源分配平台，此平台內建了監控機制，提供運行環境中所有主機與容器的資源使用情形以及服務品質數據，並可輕易地加入現有的資源分配演算法，本平台便會依據此演算法，動態地分配資源給環境中所有運行中的應用服務。

Virtualization is the essential enabling technology in cloud computing, and in recent years, there has been a rapid increase in the interest in container-based virtualization from the information technology (IT) industry. One of the most popular solutions for container-based virtualization is using Docker for container packaging with Kubernetes for multihost Docker container management. As more and more applications are deployed on the cloud, these applications running on the Kubernetes environment demand diverse resources based on different quality of service (QoS) and performance objectives. Because the resource requirement of an application can fluctuate overtime, the dynamic resource provisioning to these applications has become an important issue. Many studies have been aimed at designing a resource-provisioning algorithm to derive an optimal-resource provisioning strategy according to the QoS requirement of the applications. To allow these algorithms to be easily applied to the Kubernetes environment, this paper aims to develop a platform to realize dynamic resource provisioning based on Kubernetes.

Abstract i
Contents ii
List of Figures iii
List of Tables iv
1 Introduction 1
2 Overview of Container-based Virtualization and Docker 6
2.1 Container-based virtualization 6
2.2 Docker 7
3 The Container-based Cluster Management System 9
4 Implementation of the Dynamic Resource-provisioning Procedure 13
4.1 The Monitor Module 13
4.1.1 System resource metrics monitoring 14
4.1.2 Application performance metrics monitoring 16
4.2 The Data Aggregator Module 19
4.3 The Resource Scheduler Module 21
4.4 The Pod Scaler Module 21
5 Experimental Demonstration 25
5.1 The CPU Bound Program 26
5.2 JBoss Sample Application – Ticket Monster 27
6 Conclusion 28
7 Appendix A 29
7.1 The used InfluxQL queries 29
7.2 The Data Aggregator Output File 33

[1] Peter M. Mell and Timothy Grance. SP 800-145. The NIST Definition of Cloud
Computing. Tech. rep. Gaithersburg, MD, United States, 2011.
[2] Michael Armbrust et al. “A View of Cloud Computing”. Commun. ACM 53.4
(Apr. 2010), pp. 50–58. ISSN: 0001-0782. DOI: 10.1145/1721654.1721672.
URL: http://doi.acm.org/10.1145/1721654.1721672.
[3] Michael Armbrust et al. Above the Clouds: A Berkeley View of Cloud Computing.
Feb. 2009. URL: http://www.eecs.berkeley.edu/Pubs/TechRpts/2009/
EECS-2009-28.html.
[4] Cloud Foundry Foundation. Hope Versus Reality: Containers In 2016. Tech.
rep. June 2016. URL: https : / / www . cloudfoundry . org / learn / 2016 -
container-report/.
[5] D. Bernstein. “Containers and Cloud: From LXC to Docker to Kubernetes”.
IEEE Cloud Computing 1.3 (Sept. 2014), pp. 81–84. ISSN: 2325-6095. DOI:
10.1109/MCC.2014.51.
[6] Qi Zhang et al. “Dynamic Energy-aware Capacity Provisioning for Cloud Computing
Environments”. Proceedings of the 9th International Conference on Autonomic
Computing. ICAC ’12. ACM, 2012, pp. 145–154. ISBN: 978-1-4503-
1520-3. DOI: 10.1145/2371536.2371562. URL: http://doi.acm.org/10.
1145/2371536.2371562.
[7] D. Ardagna et al. “Energy-Aware Autonomic Resource Allocation in Multitier
Virtualized Environments”. IEEE Transactions on Services Computing 5.1 (Jan.
2012), pp. 2–19. ISSN: 1939-1374. DOI: 10.1109/TSC.2010.42.
[8] J. Bi et al. “Dynamic Provisioning Modeling for Virtualized Multi-tier Applications
in Cloud Data Center”. 2010 IEEE 3rd International Conference on Cloud
Computing. July 2010, pp. 370–377. DOI: 10.1109/CLOUD.2010.53.
[9] Docker. [Online]. Available: https://www.docker.com/.
[10] Google Cloud Platform. [Online]. Available: https://cloud.google.com/.
[11] IBM Bluemix. [Online]. Available: http://www.ibm.com/cloud-computing/
bluemix/.
[12] Microsoft Azure. [Online]. Available: https://azure.microsoft.com.
[13] kubernetes. [Online]. Available: http://kubernetes.io/.
[14] Man page of IPTABLES - Ipset - Netfilter. [Online]. Available: http://ipset.
netfilter.org/iptables.man.html.
[15] E. Casalicchio and L. Silvestri. “Architectures for autonomic service management
in cloud-based systems”. Computers and Communications (ISCC), 2011
IEEE Symposium on. June 2011, pp. 161–166. DOI: 10 . 1109 / ISCC . 2011 .
5984010.
[16] Heapster. [Online]. Available: https://github.com/kubernetes/heapster.
[17] InfluxDB. [Online]. Available: https://influxdata.com/time- series-
platform/influxdb/.
[18] Grafana. [Online]. Available: http://grafana.org/.
[19] Apache JMeter. [Online]. Available: http://jmeter.apache.org/.
[20] OpenShift Origin. [Online]. Available: https://www.openshift.org.
[21] Ticket Monster tutorial. [Online]. Available: http://developers.redhat.
com/ticket-monster/.