近年來，雲端運算成為受歡迎的電子商務服務遞送模式。現今，許多企業透過雲端運算提供電子商務服務。企業可以在特定的時間將具有應用程式服務、開發平台及硬體資源的虛擬機器(Virtual Machine, VM)資源指派給使用者。使用者可透過瀏覽器介面來購買所需要的服務(例如：應用軟體、虛擬機器及儲存空間等)，然而，使用者大多不了解服務的系統運作細節。企業利用虛擬機器提供與指派策略來解決這個問題。

虛擬機器提供與指派策略是指雲端系統提供虛擬機台並指派運算資源給使用者，以期達到服務最多的使用者之方法。然而，使用者付費之後等待一段時間，系統依然沒有開始服務使用者將會要求退費而導致企業的損失成本(loss cost)產生。損失成本可以被中途離開的使用者人數和本次使用者預計消費金額描述。這會造成企業的獲利減少。企業為提升服務品質欲達成請求雲端服務的使用者都服務，但往往企業目前擁有的計算資源在短時間內有限(因為購買實體機台或是部屬虛擬機器都需要一些時間)。使得企業必須制定排程管理策略讓使系統內中途離開使用者造成損失成本最少。有效的排程規則可以提供企業在不增加營運成本(Operation cost)的情況下有效減少企業的損失成本(loss cost)。在本論文中，損失成本是指使用者已經購買服務，由於企業提供的系統讓使用者等待不耐煩導致使用者在接收服務前離開系統並要求企業退還購買服務的金額。企業要在隨機的雲端環境中制定一個最小化損失成本的資源指派策略，決定哪種類型的使用者要先服務才能讓企業平均損失成本最小。

在許多虛擬機器提供與指派相關之文獻中，大部分研究皆針對虛擬機台的部屬數量及使用者負載的配置進行研究，相較之下，過去在使用者行為上的相關研究並不豐富，但考慮使用者行為的變化改善虛擬機台的指派方式，並且可減少企業的損失成本。目前雲端中服務導向的資源提供及指派策略可以分成兩類:(1)任務資訊驅動演算法(2)利益驅動演算法。第一部分的研究主要，雲端系統透過工作並行化來提高資源利用率，但沒有針對各個使用者等待的行為；在第二部分，雲端系統根據使用者的價值進行分配任務，造成使用者在未接受服務前中途離開系統(reneging)。本研究希望針對使用者接受服務的行為設計一個虛擬機器提供與指派策略，以期有效利用此使用者行為之優勢，達到更有效率之虛擬機器提供與指派。

我們設計一個考慮低耐性使用者的先佔方式虛擬機器提供與指派策略。本研究
主要目的是在雲端系統提供服務低耐性使用者的環境下，將可用的運算資源依照使用者等待的行為利用先佔式的方式來進行排程使用者，並達到損失成本最小化之目標。本研究首先將此問題分成三個部分來進行：在第一階段，本研究利用使用者服務資訊並設計一個使用者重要性評分將使用者分群；接著，利用不同使用者重要性評分的使用者來排程進入不同排隊佇列等待(priority queue link)，決定使用者進入服務的優先順序；最後，利用隨機派翠網數學模型來評估策略的運作方式，決定使用者先佔式指派機台的行為。所以，考慮低耐性使用者的先佔方式虛擬機器提供與指派策略主要可分成(1)分類使用者程序(2) 具優先度佇列排隊演算法(3)先佔方式虛擬機器指派演算法。在(1)分類使用者程序中，我們考慮使用者的個人資訊來進行使用者分類，進而作為指派虛擬機台的依據；在(2)具優先度佇列排隊演算法中，我們設計具優先度的排隊佇列，根據使用者影響性排名 將使用者排程到不同佇列中等待，進而提供不同品質的服務(differentiated quality of service)；在(3)先佔方式虛擬機器指派演算法，我們採用先佔式虛擬機指派，讓系統影響較大的佇列進行插隊，進而動態控制使用者排隊長度，達到降低損失成本最小化的目標。

本研究針對上述步驟進行實驗，以了解我們提出的策略所得到效能表現，並透過隨機派翠網模型找出一組最佳組合來執行先佔式指派機台的運作方法。經隨機派翠網模型分析完所得到的策略，由此可說明我們的策略方式會依照系統使用者抵達分佈及使用者等待的行為而變動，進而讓損失成本最小化之目標。此外，與一般先進先出(FIFO)的虛擬機器提供與指派策略相比，在效能方面有很大的改善。

The performance of virtual machine that provides and assigns strategies in cloud computing is high related with user’s behavior (i.e., user reneging).
In this paper, we propose Preemptive-based Strategy for Virtual Machine Provision and Allocation approach to help SaaS vendor more cost-efficient management strategies of virtual machine providing and assigning.
The main purpose of this study is to provide the services with low tolerance user in cloud system environment, then SaaS vendor assigns available resources to preempt user priorities based on user behavior, in order to minimize the loss cost.
The experimental results show that our strategy approach will arrive distribution and user behavior in accordance with the system waits for the user and change, and then minimize the loss of cost.

[1] M. Armbrust, A. Fox, R. Griffith, A. D. Joseph, R. Katz, A. Konwinski, G. Lee, D. Patterson, A. Rabkin, I. Stoica, and M. Zaharia, “A view of cloud computing,” Communications of the ACM, vol. 53, pp. 50-58, April 2010.
[2] “Abiquo,” [Online]. Available: http://www.abiquo.com/.
[3] “Accelops,” [Online]. Available: http://www.accelops.com/.
[4] “appdynamics.” [Online]. Available: http://www.appdynamics.com/.
[5] “Amazon ec2.” [Online]. Available: http://aws.amazon.com/cn/ec2/.
[6] “Appistry.” [Online]. Available: http://www.appistry.com/.
[7] “Appscale.” [Online]. Available: http://www.appscale.com/.
[8] “At&t cloud 101.” [Online]. Available: https://www.synaptic.att.com/clouduser/html/cloud101/Cloud 101 Details.htm.
[9] “Bluelock.” [Online]. Available: http://www.bluelock.com/.
[10] D. Bruneo, “A stochastic model to investigate data center performance and qos in iaas cloud computing systems,” IEEE Transactions on Parallel and Distributed Systems, vol. 25, no. 3, pp. 560-569, March 2014.
[11] R. Kumar, “Economic analysis of an m/m/c/n queuing model with balking, reneging and retention of reneged customers,” Opsearch, vol. 50, no. 3, pp. 383-403,2013.
[12] R. Buyya, C. S. Yeo, S. Venugopal, J. Broberg, and I. Brandic, “Cloud computing and emerging it platforms: Vision, hype, and reality for delivering computing as the 5th utility,” Future Generation computer systems, vol. 25, no. 6, pp. 599-616, 2009.
[13] Q. Duan, Y. Yan, and A. Vasilakos, “A survey on service-oriented network virtualization toward convergence of networking and cloud computing,” IEEE Transactions on Network and Service Management, vol. 9, no. 4, pp. 373-392, December 2012.
[14] J. Cao, K. Hwang, K. Li, and A. Y. Zomaya, “Optimal multiserver configuration for profit maximization in cloud computing,” IEEE Transactions on Parallel and Distributed Systems, vol. 24, pp. 1087-1096, 2013.
[15] D. Warneke and O. Kao, “Exploiting dynamic resource allocation for efficient parallel data processing in the cloud,” IEEE Transactions on Parallel and Distributed Systems, vol. 22, no. 6, pp. 985-997, June 2011.
[16] L. Wu, S. K. Garg, S. Versteeg, and R. Buyya, “Sla-based resource provisioning for software-as-a-service applications in cloud computing environments,” IEEE Tarnsactions on Services Computing, vol. pp, pp. 1-30, 2013.
[17] P. Mell and T. Grance, “The nist definition of cloud computing,” National Institute of Standards and Technology U.S. Department of Commerce, Tech. Rep. Special Publication 800-145, September 2011.
[18] A. Castro, V. Villagra, B. Fuentes, and B. Costales, “A flexible architecture for service management in the cloud,” IEEE Transactions on Network and Service Management, vol. 11, pp. 116 - 125, 2014.
[19] A. Movaghar, “On queueing with customer impatience until the beginning of service,” Queueing Systems, vol. 29, no. 2-4, pp. 337-350, 1998.
[20] H. Liang, L.X.Cai, D. Huang, X. Shen, and D. Peng, “An smdp-based service model for interdomain resource allocation in mobile cloud networks,” IEEE Transactions on Vehicular Technology, vol. 61, no. 5, pp. 2222-2232, Jun 2012.
[21] B. Silva, P. Maciel, J. Brilhante, and A. Zimmermann, “Geoclouds modcs: A perfomability evaluation tool for disaster tolerant iaas clouds,” in IEEE Systems Conference, 2014.
[22] L. Lei, Y. Han, and Z. Zhong, “Performance analysis of device-to-device communications with frequency reuse using stochastic petri nets,” in Wireless Networking Symposium. IEEE, 2013, pp. 6354-6359.
[23] S.-T. Yee and J. A. Ventura, “Phase-type approximation of stochastic petri nets for analysis of manufacturing systems,” IEEE Transactions on Robotics and Automation, vol. 16, pp. 318-322, 2000.
[24] Y. Dong, Y. Xia, Q. Zhu, and Y. Huang, “A stochastic approach to predict performance of web service composition,” in International Symposium on Electronic Commerce and Security. IEEE, 2009, pp. 460-464.
[25] L. Jianjie, H. Zhaohui, Y. Xuan, Z. Ran, and X. Chengan, “Analysis of process of triage in disaster rescue action using stochastic petri net,” in Industrial Control and Electronics Engineering. IEEE, 2012, pp. 111-115.
[26] B. Yang and L. Wei-Hong, “Capability evaluation of air cargo export handling system using stochastic petri net,” in Logistics Systems and Intelligent Management. IEEE, 2010, pp. 1583-1593.
[27] S. Namasivayam, “Profiting from business process outsourcing,” in IT Pro. IEEE Computer Society, 2004.
[28] “Salesforce.” [Online]. Available: http://www.salesforce.com/tw/.
[29] D. Gross, J. F. Shortle, J. M. Thompson, and C. M. Harris, Fundamentals of queueing theory. John Wiley & Sons, 2013.
[30] Q. Hu and W. Yue, Markov decision processes with their applications. Springer,2007, vol. 14.
[31] M. K. Molloy, “Performance analysis using stochastic petri nets,” IEEE Transactions on Computers, vol. 100, no. 9, pp. 913-917, 1982.
[32] F. Bause and P. S. Kritzinger, Stochastic Petri Nets. Springer, 1996.
[33] C. Hirel, B. Tuffin, and K. Trivedi, “Spnp: Stochastic petri nets. version 6.0,” in Computer Performance Evaluation.Modelling Techniques and Tools, B. Haverkort, H. Bohnenkamp, and C. Smith, Eds., vol. 1786. Springer Berlin Heidelberg, 2000, pp. 354-357.