近年來虛擬化技術越來越普遍，虛擬化技術允許一台實體機器上面可以同時擁有多台不同的虛擬機器，這些虛擬機器各自都擁有不同的作業系統，並且這些作業系統互相不受到影響，如同建立一個多台電腦的資料中心，而且虛擬化技術透過虛擬化監管系統可以方便做管理，有效的利用硬體資源，節省機房空間及電力許多優點，目前已有多數企業或產品開始導入虛擬化技術來管理資料，透過虛擬化技術可以動態管理我們的硬體資源，但是在I/O部分卻只能透過更換高頻寬的儲存裝置來改善整體的I/O效能，所以我們提出了一個在虛擬機器中的多層次的存取架構監控模組，可以即時監控每一個儲存裝置的存取狀態以及每一個虛擬機器的I/O需求，利用這些資訊搭配虛擬化監管程式分配給多層次的儲存架構來達到提升I/O整體效能的功用。

Nowadays, virtualization is a very popular technology to provide an abstract hardware emulation in many data centers. A Hypervisor is a virtualization software which can create virtual machines by utilizing the hardware resources like CPU, main memory, and storage space. Especially, the I/O performance could be a bottleneck for the virtual machines. Because a Hypervisor should properly distribute resources among virtual machines efficiently, a Hypervisor with a multitier storage system can improve the I/O performance by considering the bandwidth and capacity of low-end, mid-end, and high-end storage devices. In the thesis, we propose a monitor component in a Hypervisor with a multitier storage system by dynamically monitoring the current IOPS of each virtual machine and each storage device. With the monitor component, we can provide the current situations of each virtual machine and each storage device to make proper migration in the multitier storage system.

參考文獻
[1] "Linux container wikipedia," [Online]. Available: https://en.wikipedia.org/wiki/LXC.
[2] "VMware Homepage," VMware, [Online]. Available: http://www.vmware.com..
[3] "homepage Xen project," [Online]. Available: http://www.xenproject.org/.
[4] "KVM wikipedia," [Online]. Available: https://en.wikipedia.org/wiki/Kernel-based_Virtual_Machine.
[5] "Hyper-V wikipedia," [Online]. Available: http://en.wikipedia.org/wiki/Hyper-V.
[6] "Introduction to the Hypervisor," [Online]. Available: http://www.slideshare.net/xen_com_mgr/linuxcon-japan-13-10-years-of-xen-and-beyond.
[7] Tian Luo, Siyuan Ma, Rubao Lee, Xiaodong Zhang, Deng Liu, Li Zhou, "S-CAVE: Effective SSD Caching to Improve Virtual Machine Storage Performance," in PACT '13 Proceedings of the 22nd international conference on Parallel architectures and compilation techniques, NJ, USA, 2013.
[8] L. Lin, Y. Zhu, J. Yue and Z. C. a. B. Segee, "Hot Random Off-loading: A Hybrid Storage System With Dynamic Data," in Modeling, Analysis & Simulation of Computer and Telecommunication Systems (MASCOTS), 2011 IEEE 19th International Symposium on, Singapore, 2011.
[9] H. Shi, R. V. Arumugam and C. H. F. a. K. K. Khaing, "Optimal Disk Storage Allocation for Multitier Storage System," in APMRC, 2012 Digest, Singapore, Oct. 31 2012-Nov. 2 2012.
[10] C.-K. KANG, Y.-J. CAI, C.-H. WU and P.-C. HSIU, "A Hybrid Storage Access Framework for High-Performance," in ACM Transactions on Embedded Computing Systems (TECS) - Special Issue on Risk and Trust in Embedded Critical Systems, Special Issue on Real-Time, Embedded and Cyber-Physical Systems, Special Issue on Virtual Prototyping of Parallel and Embedded Systems, New York, NY, USA, November 2014.
[11] "storage performance," [Online]. Available: http://www.storageperformance.org/home.
[12] J. Matthews, S. Trika, D. Hensgen, R. Coulson and K. Grimsrud, "Intel® Turbo Memory: Nonvolatile disk caches in the storage hierarchy of mainstream computer systems," ACM Transactions on Storage, May 2008.
[13] S.-F. Hsiao, P.-C. Hsiu and T.-W. Kuo, "A Reconfigurable Virtual Storage Device," in IEEE ISORC, 2009.
[14] "IOmeter benchmark," [Online]. Available: http://www.iometer.org/.
[15] "IOzone benchmark," [Online]. Available: http://www.iozone.org/.