臺灣博碩士論文加值系統

今日的網際網路僅對資料提供儘可能的品質服務，若要針對提供個別資料流差別服務，必須針對每個資料流去作資訊維護的工作，如此一來，會增加高速網路中核心路由器的負載。在本篇論文中所提出的差別式網路架構下，我們提出針對頻寬和延遲提出等比例分配的機制，此機制減少維護個別資流資訊的複雜度，且能提供相同的效能。首先，我們針對每個等級的資料，利用Bloom Filter的方式預估其中所包含的活動資料流數目，並根據所預估的資料流數利用權重公平佇列的方式，動態地分配頻寬給每個等級，其次，我們所採用一種變形的RED的佇列管理機制，其能根據資料流數目和佇列長度，動態調整丟棄封包的機率，以達到穩定佇列長度的目的，因此，我們能藉由所分配的頻寬以調整佇列長度，即控制每個等級間的資料延遲，並能保證一定的延遲上限。這篇論文中整理了關於頻寬分配的相關機制，同時經由詳細的模擬，證明我們所提出的機制可以達到相當好的效能，提供各種等級資料流差別式的品質服務保證，並且適用於各種網路狀態。

In order to save network capacity there are efforts to integrate different services such as voice, video, and data. Since the different applications have particular requirements, it is necessary to provide and guarantee different levels of Quality of Service (QoS). The differentiated services concept which classifies the network traffic into different classes has the potential to save capacity and provide different levels of service.
The Assured Forwarding Per Hop Behavior (AF PHB) standardized by the Internet Engineering Task Force (IETF) Differentiated Services working group provides class-based differentiated IP services. In this relative QoS model, services obtained by different classes are proportional. However, services differentiation among classes may cause the unfairness of each flow. Therefore, we propose a dynamic Weighted Fair Queuing (WFQ) scheduling scheme to ensure the proportional fairness of each flow belong to different classes. This scheme dynamically changes the scheduler’s weights to be proportional to the estimation of active flow number of each class. In this way, services allocation can be more efficient and more feasible. By limiting the buffer size adequately, we can also guarantee proportional delay bound in the worst case.

1 Introduction 1
1.1 Internet Overview 1
1.2 Motivation for the Thesis 3
1.3 Organization 4
2 Background 5
2.1 Differentiated Services 5
2.2 Proportional QoS Model 10
2.3 Scheduling Schemes 11
2.3.1 Weighted Fair Queuing 11
2.3.2 Class-Based Queuing 12
2.3.3 Waiting Time Priority 13
2.4 Queue Management Schemes 13
2.4.1 Droptail 13
2.4.2 Random Early Detection 14
2.4.3 RED with In/Out 15
3 Providing per-Flow QoS Differentiation 17
3.1 The Proposed Architecture 17
3.2 Active Flow Number Estimation 18
3.3 Dynamic Bandwidth Allocation 19
3.4 Queue Management 21
3.4.1 Buffer Size Allocation 22
3.4.2 RED with Flow Number Estimation and Feedback Control 23
3.5 Absolute Bandwidth and Delay Guarantee 28
4 Simulation Results 29
4.1 Simulation Environment 30
4.2 Static Conditions 31
4.3 Influences of Different RTT 39
4.4 Flow Number Varies with Time 43
4.5 Minimum Bandwidth Guarantee 46
5 Conclusions 48
6 References 50

[1]R. Braden, D. Clark, and S. Shenker, “Integrated Services in the Internet Architecture: an Overview,” IETF RFC 1633, July 1994.
[2]S. Blake, D. Black, M. Carlson, E. Davies, Z. Wang, and W. Weiss, “An Architecture for Differentiated Services,” IETF RFC 2475, December 1998.
[3]C. Dovrolis, D.Stiliadis, and P. Ramanathan, “A Case for Relative Differentiated Services and the Proportional Differentiation model,” IEEE Network, September 1999.
[4]C. Dovrolis, D.Stiliadis, and P. Ramanathan, “Proportional Differentiated Services: Delay Differentiation and Packet Scheduling,” IEEE/ACM Trans. on Networking, Vol. 10, No.1, Feb. 2002.
[5]Y. Chen, M. Hamdi, H.K. Tsang, and C. Qiao, “Proportional QoS Provision: A Uniform and practical Solution,” IEEE International Conference on Communications, 2002, Vol. 4, pp. 2363 -2367, 2002.
[6]C. C. Li, S. L. Tsao, M. C. Cheng, Y. Sun, and Y. M. Huang, “Proportional Delay Differentiation Service Based on weighted Fair Queuing,” IEEE 9th International Conference on Computer Communication and Network (IC3N), , October 2000.
[7]V. Jacobson, K. Nichols, and K. Poduri, “An Expedited Forwarding PHB,” IETF RFC 2598, June 1999.
[8]J. Heinanen, T. Finland, F. Baker, W. Weiss, and J. Wroclawski, “Assured Forwarding PHB Group,” IETF RFC 2597, June 1999.
[9]C. Lin, Y. Jiang, and W. Zhou, "Integrated Performance Evaluation Criteria for Network Traffic Control," IEICE Trans. on Communications, E85-B (11): 2447-2456, November 2002.
[10]A. K. Parekh and R. G. Gallager, "A Generalized Processor Sharing Approach to Flow Control: The single Node Case," In Proceedings of IEEE INFOCOM''92, May 1992.
[11]S. Floyd and V. Jacobson, "Link-Sharing and resource management models for packet networks," IEEE/ACM Trans. on Networking, Vol.3, pp. 365-386, August 1995.
[12]S. Floyd and V. Jacobson, “Random early detection gateways for congestion avoidance,” IEEE/ACM Trans. on Networking, Vol. 1, No. 4, pp. 397-413, August 1993.
[13]D. D. Clark and W. Fang, "Explicit allocation of best-effort packet delivery services," IEEEACM Trans. on Networking, Vol.6, pp. 362-373, August 1998.
[14]B. Bloom, “Space/time Trade-offs in Hash Coding with Allowable Errors,” Communications of the ACM, 13(7), July 1970.
[15]V. Misra, W. B. Gong, and D. Towsley, “Stochastic Differential Equation Modeling and Analysis of TCP Windowsize Behavior.” Technical Report ECE-TR-CCS-99-10-01, 1999 Istanbul. Available at
ftp://gaia.cs.umass.edu/pub/Misra99-TCP-Stochastic.ps.gz
[16]J.Padhye, V. Firoiu, D. Towsley, and J. Kurose, “Modeling TCP throughput: A simple model and its empirical validation,” In Proceedings of ACM/SIGCOMM, 1998.
[17]R. Morris, “Scalable TCP Congestion Control,” In Proceedings of IEEE INFOCOM 2000, pp. 1176 — 1183.
[18]V. Misra, W. B. Gong, and D. Towsley, “Fluid-based Analysis of a Network of QUEUE MANAGEMENT Routers Supporting TCP Flows with an Application to RED,” In Proceedings of IEEE INFOCOM, 2001.
[19][Online]. Available at : http://www.aciri.org/floyed/REDparameters.txt
[20]http://www.isi.edu/nsnam/ns/