請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/83844
完整後設資料紀錄
DC 欄位 | 值 | 語言 |
---|---|---|
dc.contributor.advisor | 蔡志宏(Zse-hong Tsai) | |
dc.contributor.author | Jen-Yu Pan | en |
dc.contributor.author | 潘荏羽 | zh_TW |
dc.date.accessioned | 2023-03-19T21:20:19Z | - |
dc.date.copyright | 2022-09-30 | |
dc.date.issued | 2022 | |
dc.date.submitted | 2022-09-26 | |
dc.identifier.citation | [1] C. Joe-Wong, S. Sen, T. Lan and M. Chiang, 'Multi-resource allocation: Fairness-efficiency tradeoffs in a unifying framework,' 2012 Proceedings IEEE INFOCOM, 2012, pp. 1206-1214. [2] A. N. Toosi, R. N. Calheiros, and R. Buyya, “Interconnected cloud computing environments: Challenges, taxonomy, and survey,” ACM Comput. Surv., vol. 47, no. 1, pp. 7:1–7:47, May 2014. [3] M. Assis and L. Bittencourt, “A survey on cloud federation architectures: Identifying functional and non-functional properties,” Journal of Network and Computer Applications, vol. 72, pp. 51 – 71, 2016. [4] Amazon EC2, 2013, http://aws.amazon.com/ec2/ [5] Windows Azure Pricing Calculator, 2013, http://www.windowsazure.com/en-us/pricing/calculator/. [6] Google App Engine, 2013, https://appengine.google.com/start. [7] A. Ghodsi, M. Zaharia, B. Hindman, A. Konwinski, S. Shenker, and I. Stoica, “Dominant Resource Fairness: Fair Allocation of Multiple Resource Types,” in Proc. of USENIX NSDI, March 2011. [8] B. C. Tak, B. Urgaonkar and A. Sivasubramaniam, 'Cloudy with a Chance of Cost Savings,' in IEEE Transactions on Parallel and Distributed Systems, vol. 24, no. 6, pp. 1223-1233, June 2013. [9] R. Buyya, R. Ranjan, and R. N. Calheiros, “InterCloud: utility-oriented federation of cloud computing environments for scaling of application services,” in Proc. 10th Int. Conf. Algorithms Archit. Parallel Process., vol. 6081, pp. 13–31, May 2010. [10] S. Nesmachnow, S. Iturriaga and B. Dorronsoro, 'Efficient Heuristics for Profit Optimization of Virtual Cloud Brokers,' IEEE Computational Intelligence Magazine, vol. 10, no. 1, pp. 33-43, Feb. 2015. [11] N. Grozev and R. Buyya, “Inter-cloud architectures and application brokering: Taxonomy and survey,” Softw.-Pract. Exp., vol. 44, no. 3, pp.369-390, 2014. [12] K. M. Sim, 'Complex and Concurrent Negotiations for Multiple Interrelated e-Markets,' in IEEE Transactions on Cybernetics, vol. 43, no. 1, pp. 230-245, Feb. 2013. [13] F. Larumbe and B. Sans?, 'Green Cloud Broker: On-line Dynamic Virtual Machine Placement Across Multiple Cloud Providers,' 2016 5th IEEE International Conference on Cloud Networking (Cloudnet), 2016, pp. 119-125. [14] A. M. Manasrah, T. Smadi, and A. Almomani, “A variable service broker routing policy for data center selection in cloud analyst,” J. King Saud Univ. Comput. Inf. Sci., vol. 29, pp. 365–377, 2017. [15] Hamid Sarbazi-Azad; Albert Y. Zomaya, 'A Cloud Broker Architecture for Multicloud Environments,' in Large Scale Network-Centric Distributed Systems, IEEE, 2014, pp.359-376. [16] J. Tordsson, R. S. Montero, R. Moreno-Vozmediano, and I. M. Llorente, “Cloud brokering mechanisms for optimized placement of virtual machines across multiple providers,” Future Generation Comput. Syst., vol. 28, no. 2, pp. 358–367, 2012. [17] G. Saha and R. Pasumarthy, 'Maximizing profit of Cloud Brokers under Quantized Billing Cycles: a Dynamic pricing strategy based on ski-rental problem,' 2015 53rd Annual Allerton Conference on Communication, Control, and Computing (Allerton), 2015, pp. 1000-1007. [18] More Google cluster data, google research blog. [Online]. Available: https://research.google/tools/datasets/google-cluster-workload-traces-2019/ [19] X. Wang et al., 'Maximizing the Profit of Cloud Broker with Priority Aware Pricing,' Proc. 2017 IEEE 23rd International Conference on Parallel and Distributed Systems (ICPADS), 2017, pp. 511-518. [20] S. Jain, S. Purini and P. V. Reddy, 'A Multi-Cloud Marketplace Model with Multiple Brokers for IaaS Layer and Generalized Stable Matching,' Proc. 2018 IEEE/ACM 11th International Conference on Utility and Cloud Computing (UCC), 2018, pp. 257-266. [21] Churchill, G. A. Jr., and C. Surprenant. “An Investigation into the Determinants of Customer Satisfaction.” Journal of Marketing Research, vol. 19, no. 4, 1982, pp. 491–504, 8 Sep. 2022. [22] H. Jin, X. Wang, S. Wu, S. Di and X. Shi, 'Towards Optimized Fine-Grained Pricing of IaaS Cloud Platform,' in IEEE Transactions on Cloud Computing, vol. 3, no. 4, pp. 436-448, 1 Oct.-Dec. 2015. [23] http://ws5.cc.ntu.edu.tw/ [24] F. Fowley, C. Pahl, P. Jamshidi, D. Fang and X. Liu, 'A Classification and Comparison Framework for Cloud Service Brokerage Architectures,' IEEE Transactions on Cloud Computing, vol. 6, no. 2, pp. 358-371, 1 April-June 2018. [25] Jain, Rajendra K., Dah-Ming W. Chiu, and William R. Hawe. 'A quantitative measure of fairness and discrimination.' Eastern Research Laboratory, Digital Equipment Corporation, Hudson, MA 21 (1984). [26] Ian Jacques, “Mathematics for Economics and Business” [27] A. N. Toosi, R. K. Thulasiram and R. Buyya, 'Financial Option Market Model for Federated Cloud Environments,' 2012 IEEE Fifth International Conference on Utility and Cloud Computing, 2012, pp. 3-12. [28] J. Mei, K. Li and K. Li, 'Customer-Satisfaction-Aware Optimal Multiserver Configuration for Profit Maximization in Cloud Computing,' in IEEE Transactions on Sustainable Computing, vol. 2, no. 1, pp. 17-29, 1 Jan.-March 2017. | |
dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/83844 | - |
dc.description.abstract | 近幾年來,雲端處理的服務顯著增加。雲端計算的一個重要特徵是現付現拿。然而,受限於傳統雲端運算的市場結構,雲使用者支付的費用往往高於實際使用量,而且無法滿足雲使用者。因此,為了解決雲端運算所面臨的問題,許多研究探討如何結合一個新的參與者 - 雲代理商以更滿足典型的雲使用者的需求。雲代理商定期從雲端以批發價租用一些預留的虛擬器,並以低於雲端提供的標準價的價格提供給雲使用者。此外,在這個市場中,與雲供應商相比,雲代理商通常可採用更短的計費周期,因此雲代理商可以為雲使用者降低大量時間成本。 然而,大多數以雲代理商為中心的研究都集中在如何提高市場表現,例如產生更多利潤或最小化成本消耗。在這些研究中很少有人考慮到真實市場的限制,即市場中雲供應商的虛擬器數量有限。由於雲供應商的虛擬器昂貴且有限,雲代理商將接收大量雲使用者的需求,尤其是在大型市場中。 此篇論文在以雲代理商為中心的市場架構下,我們提出一種新的以雲代理商為中心的虛擬器分配演算法。為了解決虛擬器數量有限的問題,我們採用所謂的信用評分規則和定價規則來決定虛擬器分別由雲供應商和雲代理商租出和購買的數量和價格。雲代理商的運作又可以分為兩種模式:以利潤為目標的模式和在以公平為目標的模式。利潤目標模式側重於如何配置虛擬器以及如何為其定價,以為雲代理商之用戶節省成本的前提下最大化其利潤,其中雲代理商的利潤受許多因素影響,像是使用者的需求、虛擬器的買賣價格、市場的規模等等。另一個公平目標模式側重於滿足用戶需求,在這種模式下,雲代理商的目標不是關注自身的利潤,而是要在使用者的虛擬器分配上實現公平。 此篇論文以雲代理商模型實作市場架構,並評估論文所提出的演算法在不同環境下的效能。結果顯示,論文提出的這種新的以雲代理商為中心的多雲環境被證明是有效的,即使在使用者需求波動很大的多雲環境下也是如此。此外,提出的虛擬器分配演算法可以有效地為使用者降低成本,在資源短缺的情況下有效地保持公平,同時通過不同的商業運作模式獲得可觀的利潤。 | zh_TW |
dc.description.abstract | Over the past few years, cloud computing has experienced remarkable development. One important feature of cloud computing is pay-as-you-go. However, users often pay more than their actual usage due to the current cloud market structure. In this situation, traditional cloud providers may not able to satisfy the QoS of end users because of high price. Therefore, we introduce a new stakeholder to better meet typical cloud user demands, which is the service broker in the cloud market. A service broker periodically leases a number of reserved instances from cloud providers with a whole-sale price and then offers them to users on an on-demand basis at a price cheaper than the standard price of cloud providers. Also, in this market the broker usually adopts a shorter billing cycle compared with cloud providers. By doing this, the broker can reduce a great amount of cost for users. However, most broker-centric researches focused on how to improve the performance of the market such as generating more profit or minimizing cost consumption. Few of them considered the physical limitation in the real market, that is, the limited number of instances of cloud providers in Broker-Centric Market. That is, because the number of instances in cloud providers is expensive and limited, service broker will accommodate a large number of end users especially in large scale marketplace. In this thesis, we propose a novel broker-centric instance-allocating algorithm for Broker-Centric marketplace. I n order to address the problem of limited number of instances, we adopt the so-called credit rules and pricing rules to determine the number and the price of instances to lease and purchase by cloud providers and service brokers respectively. The business operation of this service broker can further be divided into two modes: the profit-objective mode and the fairness-objective mode. The profit-objective mode focuses on how to configure a broker and how to price its instances such that its profit can be maximized on the premise of saving costs for users. Profit of a broker is affected by many factors such as the user demands, the purchase price, the sales price of instances, and the scale of the broker, etc. The fairness-objective mode, in addition, is activated when the total number of instances is too less to meet the users’ demands, and the broker wants to satisfy user requirements as much as it can. In this mode, instead of focusing on the profit of the broker, the objective of the broker is to achieve minimax fairness on instance allocation for user applications to assure virtual machines resources are fairly allocated. In the final part of this thesis, the performance of Broker-Centric marketplace and Broker-Centric instance-allocating algorithm are validated via a series of experiments. This new broker-centric cloud market system is proved to be effective even when working on a large-scale cloud marketplace even with high fluctuation of users’ demands. Moreover, this system also shows that the broker can efficiently reduce the cost for users, and effectively maintain fairness upon resource shortage, while still making a considerable profit via different modes of business operations. | en |
dc.description.provenance | Made available in DSpace on 2023-03-19T21:20:19Z (GMT). No. of bitstreams: 1 U0001-2609202218315700.pdf: 3104641 bytes, checksum: c04cd576842f603835d7cd11c2bceb70 (MD5) Previous issue date: 2022 | en |
dc.description.tableofcontents | 誌謝 i 中文摘要 ii ABSTRACT iv CONTENTS vi LIST OF FIGURES viii LIST OF TABLES x Chapter 1 Introduction 1 1.1 Introduction 1 1.2 Related Work 2 1.3 Motivation 6 Chapter 2 Multi-Cloud Marketplace Structure 8 2.1 System Architecture 8 2.2 Multi-Cloud Marketplace Model 11 2.3 Market Operation 18 2.3.1 Various Kinds of Tasks by End Users 25 2.3.2 User Satisfaction and Social Welfare 27 2.3.3 Credit Rules and Instance Supply Decision in Cloud Provider 29 2.3.4 Rule for the Pricing of the Service Broker and Cloud Provider 32 Chapter 3 Broker-Centric Marketplace Instance-Allocating Algorithm 34 3.1 The Broker-Centric Instance-Allocating Algorithm 34 3.2 Profit-Objective Optimization 35 3.3 Fairness-Objective Optimization 39 Chapter 4 Simulation and Analysis 44 4.1 Simulation Setup 46 4.2 Parameters 47 4.3 Experiment 1 – Evaluate Performance with the Window Size Factor 'x” 49 4.4 Experiment 2 – Evaluating Performance Impact from Different Factors 53 4.4.1 Scenario 2.1 – Market Behavior under Time Varying User Demand 53 4.4.2 Scenario 2.2 – Market Behavior with Different Number of Available Cloud Instances 63 4.5 Experiment 3 – Broker with Extreme Behavior 66 4.6 Experiment 4 – Performance under Real-World Traces 70 4.7 Summary Remarks of Simulation Results 73 Chapter 5 Conclusions 75 5.1 Conclusions 75 5.2 Future Works 76 Bibliography 77 | |
dc.language.iso | en | |
dc.title | 多雲環境中雲代理商獲利最佳化與近似公平之市場模型 | zh_TW |
dc.title | A Broker-Centric Multi-Cloud Marketplace Model for Profit Optimization with Near Minimax Fairness | en |
dc.type | Thesis | |
dc.date.schoolyear | 110-2 | |
dc.description.degree | 碩士 | |
dc.contributor.oralexamcommittee | 林風(P Lin),鍾耀梁(Yao-Liang Chung) | |
dc.subject.keyword | 雲端計算,運算資源分配,經濟學,虛擬器,最佳化, | zh_TW |
dc.subject.keyword | Cloud computing,Pricing,Economics,Virtual machine,Optimization, | en |
dc.relation.page | 80 | |
dc.identifier.doi | 10.6342/NTU202204111 | |
dc.rights.note | 未授權 | |
dc.date.accepted | 2022-09-27 | |
dc.contributor.author-college | 電機資訊學院 | zh_TW |
dc.contributor.author-dept | 電信工程學研究所 | zh_TW |
顯示於系所單位: | 電信工程學研究所 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
U0001-2609202218315700.pdf 目前未授權公開取用 | 3.03 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。