雲端運算(Cloud Computing)是描述一種基於網際網路(Internet)的服務，雲端運算不是一種新技術，而是一種虛擬概念的代表，其目的在動態調整資源配置以減少建置成本、提高資源使用率及服務效能、並提供更強的系統備援系統管理自動化等。雲端運算其實就是指網路運算與網路存儲存的概念，其本質承襲自分散式運算(Distributed Computing)。如我們使用的Google的服務，就不會僅是一個超級電腦，而是由一大群電腦所組成的雲端運算，提供所有的服務。雲端運算不是只有分散式運算，還包括大量資料的儲存、備份、與存取。
近年來，在雲端資料存取網路(Storage Area Network, SAN)中，TCP協定特性的關係，被發現其產生的Incast現象影響雲端資料處理效能甚鉅，已有一些文獻提出解決方案。其中Hashed Credits Fair方法提出透過credit與高低優先權佇列方式以平衡各資料流的傳輸量來解決Incast問題。本研究發現其方式有可能造成封包錯序的問題，因此本論文除探入討論其錯序現象外，也提出解決方案，除此之外，並從實驗模擬中觀察到所提方法在高低優先權佇列的使用也比既有方法更為平均。

Cloud computing, which is not a new technology, a kind of internet services. Basically, it is a concept of virtual concept to flexibly reconfig the network respources to decrease installation cost, increase resource utilization and service performance, data backup and automatic management. Cloud computing is naturally a distributed processing concept and it includes computing and storage management. The virtualization makes the Recently, it was found that, due to the TCP protocol characteristics, the existence of incast issue in the storage area network greatly affects the processing performance of the cloud computing. There are several researches proposed to deal with this issue. Among them, the hashed credits fair (HCF) scheme applies the concept of credit and high/low priority queues to balance the data flows so as to resolve the incast problem. However, we found that the HCF scheme may introduce the out of sequence (OOS) problem. In this thesis, we study and analyze the OOS issue in HCF and propose a scheme to solve it. In addition, we also investigate the queuing behaviors through exhaustive simulations and find that the proposed scheme shows more balance in utilizing the high and low priotity queues.

[1] Buyya, Rajkumar; Chee Shin Yeo, Srikumar Venugopal. Market-Oriented Cloud Computing: Vision, Hype, and Reality for Delivering IT Services as Computing Utilities (PDF). Department of Computer Science and Software Engineering, The University of Melbourne, Australia: 9 [2008-07-31].
[2] The Software-RAID HOWTO, Jakob Østergaard and Emilio Bueso, v.1.1.1 2010-03-06, http://www.tldp.org/HOWTO/Software-RAID-HOWTO.html
[3] RFC-3720 - Internet Small Computer Systems Interface (iSCSI)
[4] David Nagle, Denis Serenyi and Abbie Matthews. “The Panasas activescale storage cluster: Delivering scalable high bandwidth storage”. In SC ’04: Proceedings of the 2004 ACM/IEEE conference on Supercomputing, page 53, Washington, DC, USA, 2004. IEEE Computer Society.
[5] ANSI SCSI Standard, X3.131-1986, June 23, 1986, 2nd, foreword
[6] RFC-1094 Network File System Protocol Specification
[7] IEEE Standards 802.3, IEEE Stand for Information technology - Telecommunications and information exchange between systems - Local and metropolitan area networks Specific requirements Part3. Carrier sense multiple access with collision detection (CSMA/CD) access method and physical layer specifications., IEEE Computer Society, 2005。
[8] ANSI X3.230-1994, (FC-PH). Information Technology - Fiber Channel - Physical and Signaling Interface
[9] Paul Shread (April 10, 2008). "Cisco Buys Nuova as FCoE Heats Up". Enterprise Storage Forum. Retrieved May 5, 2011.
[10] Gruman, Galen. What cloud computing really means. InfoWorld. 2008-04-07 [2009-06-02].
[11] 作者：陳彥文、劉惠英、許獻聰, 網路交換技術, 出版社：全華圖書 出版日期：2007年09月13日
[12] A. Phanishayee, E. Krevat, V. Vasudevan, D. G. Andersen, G. R. Ganger, G. A. Gibson and S. Seshan, “Measurement and analysis of TCP throughput collapse in cluster-based storage systems,” FAST ’08, Feb. 2008, San Jose, CA.
[13] Y. Chen, R. Griffith, J. Liu, R. H. Katz and A. D. Joseph, “Understanding TCP incast throughput collapse in datacenter networks,” WREN’09, Aug. 21, 2009, Barcelona, Spain.
[14] V. Vasudevan, A. Phanishayee, H. Shah, E. Krevat, D. G. Andersen, G. R. Ganger, G. A. Gibson and B. Mueller, “Safe and effective fine-grained TCP retransmissions for datacenter communication,” ACM SIGCOMM’09, Aug. 17-21, 2009, Barcelona, Spain.
[15] Vijay Shankar Rajanna, Smit Shah, Anand Jahagirdar, and Kartik Gopalan,” XCo: Explicit Coordination for Preventing Congestion in Data Center Ethernet”, Computer Science, State University of New York, Binghamton, NY
[16] Alexander Shpiner and Isaac Keslassy,”A Switch-Based Approach to Throughput Collapse and Starvation in Data Centers”, 16-18 June, (2010) , Technion - Israel Institute of Technology
[17] Anvin, H. Peter (21 May 2009). "The mathematics of RAID-6". Retrieved November 4, 2009.
[18]802.1Qau: IEEE Standard for Local and Metropolitan Area Networks: Virtual Bridged Local Area Networks - Amendment 10: Congestion Notification. http://www.ieee802.org/1/pages/802.1au.html
[19] The Data Center Bridging (DCB) Task Group (TG) is a part of the IEEE 802.1 Working Group. http://www.ieee802.org/1/pages/dcbridges.html
[20] RFC – 5661 Network File System (NFS) Version 4 Minor Version 1 Protocol.
[21] Preston, W. Curtis (2002). "Fibre Channel Architecture". Using SANs and NAS. Sebastopol, CA: O''Reilly Media. pp. 19–39. ISBN 978-0-596-00153-7. OCLC 472853124.
[22] Riabov, Vladmir V. (2004). "Storage Area Networks (SANs)". In Bidgoli, Hossein. The Internet Encyclopedia. Volume 3, P-Z. Hoboken, NJ: John Wiley & Sons. pp. 329–338. ISBN 978-0-471-68997-3. OCLC 55610291
[23] IEEE P802.3ba 40Gb/s and 100Gb/s Ethernet Task Force. official web site. IEEE. June 19, 2010. Retrieved June 24, 2011.