過去十年間，網路技術的蓬勃發展衍生出雲端運算的新興概念，雲端運算相關的應用程式一直是近年企業積極擴展的主力技術。資料中心是雲端技術中不可或缺的核心部分，其提供了大量儲存與運算能力並透過高頻寬、低延遲的網路環境來快速地處理與回覆使用者提出的各類需求。然而，在資料中心網路內部的獨特短時間多對一的通訊模式Incast 造成TCP 協定在效能上產生瓶頸。近年雖然有大量研究成果提出來解決此問題，但鮮少研究能夠在低成本的情況下提出一個TCP傳輸層的解決方案。
本論文中，作者提出以延遲變異為基礎的TCP協定運作於資料中心網路上 (JDCTCP)，提供精準網路壅塞的預警，進而能夠讓TCP傳送端提早並且適當的降低傳輸速率且在維持TCP基本傳輸效能的情況下來避免Incast通訊模式所導致的TCP 逾時。經由模擬的實驗觀察可證實，本論文所提出的方法JDCTCP能夠在有效的網路吞吐量(Goodput)、平均需求回應延遲時間 (Average query delay)上優於目前較著名的TCP協定，如NewReno, SACK, Westwood, DCTCP。同時JDCTCP傳送端能夠平穩的傳輸資料，以長期的網路傳輸行為觀察可知，傳送端之間也能夠公平的共享網路資源。

In the last decade, various cloud-based applications are proposed to meet the user’s requirements. Data center is the core component of the cloud-based application. It provides large-scale storage and computing resource to response the client’s request immediately over high-bandwidth and low-delay links. However, a unique incast communication pattern on data center causing the performance degradation to the commodity TCP variants. Unfortunately, few solutions can solve the problem fundamentally at low cost and based on TCP-level solution.
In this thesis, we proposed JDCTCP, a jitter-based TCP for data center networks. JDCTCP extends the concept of a jitter ratio to aware the immediate queue condition and adjust the congestion window size properly to avoiding TCP timeout. JDCTCP only have to modify the sender side which means this is an end-to-end and TCP-level solution at very low cost compare to other solutions. The simulation results reveal that JDCTCP not only maintain the stability, fairness properies, but also conduct better goodput and delay performance than popular TCP variants such as Newreno, Westwood, SACK and DCTCP.

[1] J. Postel, Transmission Control Protocol: RFC Editor, 1981.
[2] A. Phanishayee, E. Krevat, V. Vasudevan et al., “Measurement and analysis of TCP throughput collapse in cluster-based storage systems,” in Proceedings of the 6th USENIX Conference on File and Storage Technologies, San Jose, California, 2008, pp. 1-14.
[3] Y. Chen, R. Griffith, J. Liu et al., “Understanding TCP incast throughput collapse in datacenter networks,” in Proceedings of the 1st ACM workshop on Research on enterprise networking, Barcelona, Spain, 2009, pp. 73-82.
[4] P. Devkota, and A. L. N. Reddy, “Performance of Quantized Congestion Notification in TCP Incast Scenarios of Data Centers,” in Proceedings of the 2010 IEEE International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication Systems, 2010, pp. 235-243.
[5] H. Wu, Z. Feng, C. Guo et al., “ICTCP: Incast Congestion Control for TCP in data center networks,” in Proceedings of the 6th International COnference, Philadelphia, Pennsylvania, 2010, pp. 1-12.
[6] Z. Jiao, R. Fengyuan, and L. Chuang, "Modeling and understanding TCP incast in data center networks." pp. 1377-1385.
[7] V. Vasudevan, A. Phanishayee, H. Shah et al., “Safe and effective fine-grained TCP retransmissions for datacenter communication,” in Proceedings of the ACM SIGCOMM 2009 conference on Data communication, Barcelona, Spain, 2009, pp. 303-314.
[8] M. Alizadeh, A. Greenberg, D. A. Maltz et al., “Data center TCP (DCTCP),” in Proceedings of the ACM SIGCOMM 2010 conference, New Delhi, India, 2010, pp. 63-74.
[9] M. Podlesny, and C. Williamson, “An application-level solution for the TCP-incast problem in data center networks,” in Proceedings of the Nineteenth International Workshop on Quality of Service, San Jose, California, 2011, pp. 1-3.
[10] M. Gusat, R. Birke, and C. Minkenberg, “Delay-based cloud congestion control,” in Proceedings of the 28th IEEE conference on Global telecommunications, Honolulu, Hawaii, USA, 2009, pp. 5691-5698.
[11] M. Alizadeh, A. Javanmard, and B. Prabhakar, “Analysis of DCTCP: stability, convergence, and fairness,” in Proceedings of the ACM SIGMETRICS joint international conference on Measurement and modeling of computer systems, San Jose, California, USA, 2011, pp. 73-84.
[12] S. Floyd, and V. Jacobson, “Random early detection gateways for congestion avoidance,” IEEE/ACM Trans. Netw., vol. 1, no. 4, pp. 397-413, 1993.
[13] K. Ramakrishnan, S. Floyd, and D. Black, The Addition of Explicit Congestion Notification (ECN) to IP: RFC Editor, 2001.
[14] E. H.-K. Wu, and M.-Z. Chen, “JTCP: jitter-based TCP for heterogeneous wireless networks,” IEEE J.Sel. A. Commun., vol. 22, no. 4, pp. 757-766, 2006.
[15] S. Floyd, T. Henderson, and A. Gurtov, The NewReno Modification to TCP''s Fast Recovery Algorithm: RFC Editor, 2004.
[16] S. Mascolo, C. Casetti, M. Gerla et al., “TCP westwood: Bandwidth estimation for enhanced transport over wireless links,” in Proceedings of the 7th annual international conference on Mobile computing and networking, Rome, Italy, 2001, pp. 287-297.
[17] S. McCanne, and S. Floyd. "ns Network Simulator," http://www.isi.edu/nsnam/ns/.
[18] C. Guo, G. Lu, D. Li et al., “BCube: a high performance, server-centric network architecture for modular data centers,” SIGCOMM Comput. Commun. Rev., vol. 39, no. 4, pp. 63-74, 2009.
[19] A. Greenberg, J. R. Hamilton, N. Jain et al., “VL2: a scalable and flexible data center network,” SIGCOMM Comput. Commun. Rev., vol. 39, no. 4, pp. 51-62, 2009.
[20] R. N. Mysore, A. Pamboris, N. Farrington et al., “PortLand: a scalable fault-tolerant layer 2 data center network fabric,” SIGCOMM Comput. Commun. Rev., vol. 39, no. 4, pp. 39-50, 2009.