由於網路頻寬技術愈臻成熟、雲端裝置選擇愈來愈多樣化，如：電腦、智慧型手機、電視、照相機…等，使得雲端產業在近幾年蓬勃發展。雲端資料中心可視為整個雲端運算的心臟所在，其提供了有關處理、運算以及儲存…等相關服務，因此，當使用者愈多，能源消耗也將會愈高，故，如何有效率的運用資源避免不必要的資源浪費以導致能源消耗過高為首要關切之議題。
雲端運算(Cloud Computing)最大優勢為其具有實體主機虛擬化(virtualization)與虛擬主機遷移技術(migration)：運行在使用過低實體主機的虛擬主機能夠透過遷移技術遷移至其他主機，實體主機歸功於虛擬化技術能使多台虛擬主機運行於其上，透過隔離技術安全共享其所提供之資源，並將使用過低或者閒置的主機切換至低耗能模式以達成資源有效運用以及節省不必要的能源消耗之目的，此兩技術總稱為整合技術(consolidation)。
在此篇論文當中，我們提出為雲端資料中心所設計之高效率虛擬機器調控機制：有效使用雲端資源下，在能源消耗與服務品質(QoS)之間做權衡的啟發式演算法。我們以多面向評估標準與以使用虛擬主機歷史資訊(MIPS)為基礎之統計預測方法-LrMmt做比較，實驗結果得知，我們所提出之方法除了比LrMmt使用更有效率外，我們所提出之整合政策更能減少虛擬主機遷移、減少虛擬主機遷移的成本，從綜觀而言，我們提出之方法，更能夠滿足雲端資料中心大部分使用者之需求。
此外，我們也考慮另一種情境：替虛擬機加入優先權以考慮運行於虛擬主機的應用程式重要性。當應用程式執行工作越重要時，在虛擬主機遷移政策下，應選擇其他相對執行工作較不重要之虛擬主機。我們分別實驗三種情況，根據虛擬主機優先權等級分配不同權重，此權重代表著虛擬主機遷移需花費的代價，相對的我們也做了與以統計為基礎的整合方法LrMmt比較[3]，實驗結果顯示，不管是否考慮虛擬機優先等級與否，我們所提出的方法效果皆優於LrMmt。

Network technology is more and more maturity and the diversity of the computing devices used to connect to Cloud environments, such as computers, laptops, smartphones, TVs and cameras… and so on. In this condition, the Cloud industry boomed in recent years. Cloud data centers can be regarded as the heart of the entire Cloud computing, it provides computing, processing and storage services. When the users are more, the energy consumption will be higher. Therefore, how to avoid unnecessary waste of resources which leading to high energy consumption is the problem that we need to concern.
The greatest advantages of the Cloud Computing are the virtualization from the physical host and the virtual machine migration technology: Using the migration technology, a VM (i.e. virtual machine) which running through the underutilization host moves to the other one. Attributing to virtualization technology, a host enables to run multiple VMs residing on it and VMs share resources by isolation technology. To achieve the purpose of effective using resources and saving unnecessary energy consumption by turning the underutilization host to the low-power mode. Combing with this two techniques are referred to as consolidation technology.
In this thesis, we proposed the efficient virtual machine provision mechanism for Cloud data centers: it is a kind of heuristic algorithm. Besides, using Cloud resources efficiently and caring about the trade-off between the energy consumption and quality of services (QoS). In addition to present the multi-faceted evaluation criteria and comparing with the algorithm which using statistical forecasting method based on using the historical information (MIPS) of the VMs, called LrMmt. From the experiment results showed that our proposed method not only using resources more efficiently, but also less number of VM migrations and less number of SLA violations in the Cloud data centers.
In addition, we also consider another scenario: giving the priority to the VMs in order to considering the importance of the applications running on the VM. Choosing a lowest priority of VM on a host and minimum migration time of VM by the new VM selection policy. We used three cases for assign different migration overhead and compared with LrMmt. The experiment results shown that whether considering priority or not, our consolidation policy are better than LrMmt.

中文摘要
Abstract
致謝
Contents
List of Figures and Tables
Chapter 1 Introduction
Chapter 2 Related Work
2.1 Factors of Affecting Power Consumption
2.1.1 One-Dimensional: CPU
2.1.2 One-Dimensional: Memory
2.1.3 Multi-Dimensional: CPU, Memory, Disk and other else
2.2 Power Saving Considerations
2.2.1 Trade-Off Between Migration and SLA Violation
2.2.2 Only Considering Migration
2.3 The statistical method - LrMmt
Chapter 3 Efficient Provision Mechanism for Cloud Data Centers Resources
3.1 System Model
3.1.1 Power Model
3.1.2 Migration Cost Model
3.1.3 SLA Violation Model
3.2 The Virtual Machine Consolidated Policy
3.2.1 Overloading Detection Scheme
3.2.2 VM Selection Scheme
3.2.3 VM Placement Scheme
3.3 Considering Priority of Virtual Machines
Chapter 4 Simulation and Results
4.1 Experiment Setup
4.2 Performance Metrics
4.3 Workload Data
4.4 Simulation Results and Analysis
4.4.1 Optimization of Resource Consolidate Policy
4.4.2 All VMs with the Same Priority
4.4.3 Considering different priorities of VMs
Chapter 5 Conclusion
References

[1]KS. Park and V.S. Pai “CoMon: a mostly-scalable monitoring system for PlanetLab.” ACM SIGOPS Operating Systems Review, Vol. 40, NO.1, pp. 65-74, Jan 2006.
[2]X. Fan, W. Weber and L. A. Barroso “Power provisioning for a warehouse-sized computer” ISCA '07 Proceedings of the 34th annual international symposium on Computer architecture, 2007
[3]A. Beloglazov, J. and R. Buyya. “Optimal online deterministic algorithms and adaptive heuristics for energy and performance efficient dynamic consolidation of virtual machines in Cloud data centers” Vol. 24, Issue 13, pp. 1397–1420, 10 Sep. 2012
[4]C. C. Lin, P. Liu and J. J. Wu “Energy-efficient Virtual Machine Provision Algorithms for Cloud Systems” Fourth IEEE International Conference on Utility and Cloud Computing (UCC), 2011.
[5]A. Verma, P. Ahuia and A. Neoqi “pMapper: Power and Migration Cost Aware Application Placement in Virtualized Systems” Proceedings of the 9th ACM/IFIP/USENIX International Conference on Middleware, 2008
[6]A. Gandhi, M. Harchol-Balter, R. Das and C. Lefurgy “Optimal power allocation in server farms” Proceedings of the 11th International Joint Conference on Measurement and Modeling of Computer Systems, 2009.
[7]D. Kusic, J.O. Kephart, J.E. Hanson, N. Kandasamy and G. Jiang, “Power and performance management of virtualized computing environments via lookahead control”, International Journal of Cluster Computing Vol.12, No.1,pp.1-15, 2009
[8]D. Kliazovich, P. Bouvry, Y. Audzevich and S. U. Khan, “GreenCloud: a packet-level simulator of energy-aware Cloud computing data centers” International Conference on Global Telecommunications(GLOBECOM),2010
[9]A. Beloglazov, J. Abawajy and R. Buyya. “Energy-aware resource allocation heuristics for efficient management of data centers for Cloud computing” IEEE/ACM International Conference on Cluster, Cloud and Grid Computing (CCGrid), 2010.
[10]Q. Chen, P. Grosso, K. Veldt, C. Laat, R. Hofman and H. Bal “Profiling energy consumption of VMs for green Cloud computing” Proceedings of the 2011 IEEE 9th International Conference on Dependable, Autonomic and Secure Computing (DASC), 2011
[11]B. P. Rimal, E. Choi and I. Lumb “A Taxonomy and Survey of Cloud Computing Systems” Proceedings of the Fifth International Joint Conference on INC, IMS and IDC, 2009
[12]T. Heath, A. P. Centeno, P. George, L. Ramos, Y. Jaluria, and R. Biachini “Mercury and Freon: Temperature Emulation and Management in Server Systems.” International Conference on Architectural Support for Programming Languages and Operating Systems, Oct. 2006.
[13]A. E. H. Bohra and V. Chaudhary “VMeter: Power Modelling for Virtualized Clouds” IEEE International Symposium on Parallel & Distributed Processing, Workshops and Phd Forum (IPDPSW), 2011.
[14]A. Beloglazov, J. Abawajy and R. Buyya. “Adaptive threshold-based approach for energy-efficient consolidation of virtual machines in cloud data centers” Proceedings of the 8th International Workshop on Middleware for Grids, Clouds and e-Science, 2010
[15]Q. Huang, F. Gao, R. Wang and Z. Qi “Power Consumption of Virtual Machine Live Migration in Clouds” Third International Conference on Communications and Mobile Computing (CMC), 2011
[16]W. Xiaoli and L. Zhanghui “An Energy-Aware VMs Placement Algorithm in Cloud Computing Environment” International Conference on Intelligent System Design and Engineering Application (ISDEA), 2012
[17]R.N. Calheiros, R. Ranjan, A. Beloglazov, C.A.F.D. Rose and R.Buyya “CloudSim: a toolkit for modeling and simulation of Cloud computing environments and evaluation of resource provisioning algorithms. Software: Practice and Experience” International Journal of Software-Practice & Experience, Vol.41, NO.1, pp. 23-50, Jan 2011.