基于源端的ATM网络ABR业务流量控制
|
摘要
高速计算机通信网络的关键技术之一是异步传输模式（ATM：Asynchronous Transfer Mode）技术。在ATM网络中，信息的拥塞及丢失是影响网络业务服务质量的主要原因。自适应比特率（ABR：Available Bit Rate）业务是唯一一种可采用反馈机制进行流量控制的业务，因此，近年来ABR业务流量的控制和管理问题成为一个研究热点。通信网络是一个庞大的复杂系统，ATM网络拥塞控制的研究对网络的控制和通信均具有重要的理论意义和实用价值。本文正是以此为研究对象，将控制理论概念引入到网络通信中，很好地解决可控流的拥塞控制问题。
本文首先介绍了ATM网络的基本原理，ABR业务的反馈机制，以及ABR流量控制机制，分析了现有的ATM网络拥塞控制原理。
其次，对现有基于速率的流量控制算法进行了分析，这些算法中大多都能提高活动连接和连接装置的公平性，但是，也给交换机带来更大的复杂性。基于这点，本文提出了一种流量控制算法，并将部分速率计算工作从交换机转移到源端系统中进行。结果表明，该算法能够减少交换机的速率计算工作以及交换复杂性。而且，还降低了交换机在每一时间间隔内为每条链路计算负载因子的难度，从而使网络中的队列长度保持在稳定状态下，进而达到了网络的全局最优性。
最后，本文对提出的算法进行了仿真实验。由于本算法引入三个参数，这三个参数的值对本算法进行拥塞控制的效果起到决定性作用，因此，在仿真中，重点对三个参数的取值范围进行仿真，给出了该参数的最优解，同时，在相同的仿真模型中，对ERICA+算法也进行仿真，对仿真结果进行比较，结果表明该算法达到了很好的效果。该方法和算法的提出，对解决ATM网络的拥塞问题提供了理论指导，具有一定理论意义和实践意义。
One of future high-speed computer communication network technologies is Asynchronous Transfer Mode(ATM)technology. The loss and congestion of information are main reasons that affect the quality of service in ATM network. Available Bit Rate(ABR)service is the only one type of traffic that can be controlled using the feedback mechanism, so, in the recent, the control and management of ABR service becomes a hot subject. The communication network is a large and hybrid system, the research on congestion control of ATM network has great significance in theory and practice for the control and communications of network. The paper is just based on it and the control theory is used to solve the congestion of controllable flow in communications network.
The paper begins with the basic work principle of networks, and the feedback mechanism of ABR, and the flow control mechanism of ABR, and analyses ATM network congestion control principle in existence.
Secondly, analyses rate-based flow control arithmetics in existence, most of these arithmetics improve both the fairness among active connections and the link utilization by putting more and more complexity into a switch. In the paper, we propose a flow control arithmetic in which part of rate-calculation work is moved from switches to source-systems. As a result, the proposed arithmetic reduces the rate-calculation effort and switch complexity in the switch. Furthermore, it reduces the difficulty for calculating load factor every measurement interval in switches, as well as more and more stable queue occupancy in the network, thereby it attain the full furthest excellence.
In the end, we simulate for the proposed arithmetic. For three parameters put into the proposed arithmetic, they are most important for the congestion control impact of the proposed arithmetic, so in simulation, the emphases puts value scope of three parameters to simulate, and the right value is given. In the same simulation configuration, puts ERICA+ to simulate, and compares
simulation result, the result indicates that method attains good purpose. The proposed method and arithmetic affords the theory guidance for the congestion of the ATM network, and have some theory and practice significance.
本文首先介绍了ATM网络的基本原理，ABR业务的反馈机制，以及ABR流量控制机制，分析了现有的ATM网络拥塞控制原理。
其次，对现有基于速率的流量控制算法进行了分析，这些算法中大多都能提高活动连接和连接装置的公平性，但是，也给交换机带来更大的复杂性。基于这点，本文提出了一种流量控制算法，并将部分速率计算工作从交换机转移到源端系统中进行。结果表明，该算法能够减少交换机的速率计算工作以及交换复杂性。而且，还降低了交换机在每一时间间隔内为每条链路计算负载因子的难度，从而使网络中的队列长度保持在稳定状态下，进而达到了网络的全局最优性。
最后，本文对提出的算法进行了仿真实验。由于本算法引入三个参数，这三个参数的值对本算法进行拥塞控制的效果起到决定性作用，因此，在仿真中，重点对三个参数的取值范围进行仿真，给出了该参数的最优解，同时，在相同的仿真模型中，对ERICA+算法也进行仿真，对仿真结果进行比较，结果表明该算法达到了很好的效果。该方法和算法的提出，对解决ATM网络的拥塞问题提供了理论指导，具有一定理论意义和实践意义。
One of future high-speed computer communication network technologies is Asynchronous Transfer Mode(ATM)technology. The loss and congestion of information are main reasons that affect the quality of service in ATM network. Available Bit Rate(ABR)service is the only one type of traffic that can be controlled using the feedback mechanism, so, in the recent, the control and management of ABR service becomes a hot subject. The communication network is a large and hybrid system, the research on congestion control of ATM network has great significance in theory and practice for the control and communications of network. The paper is just based on it and the control theory is used to solve the congestion of controllable flow in communications network.
The paper begins with the basic work principle of networks, and the feedback mechanism of ABR, and the flow control mechanism of ABR, and analyses ATM network congestion control principle in existence.
Secondly, analyses rate-based flow control arithmetics in existence, most of these arithmetics improve both the fairness among active connections and the link utilization by putting more and more complexity into a switch. In the paper, we propose a flow control arithmetic in which part of rate-calculation work is moved from switches to source-systems. As a result, the proposed arithmetic reduces the rate-calculation effort and switch complexity in the switch. Furthermore, it reduces the difficulty for calculating load factor every measurement interval in switches, as well as more and more stable queue occupancy in the network, thereby it attain the full furthest excellence.
In the end, we simulate for the proposed arithmetic. For three parameters put into the proposed arithmetic, they are most important for the congestion control impact of the proposed arithmetic, so in simulation, the emphases puts value scope of three parameters to simulate, and the right value is given. In the same simulation configuration, puts ERICA+ to simulate, and compares
simulation result, the result indicates that method attains good purpose. The proposed method and arithmetic affords the theory guidance for the congestion of the ATM network, and have some theory and practice significance.
引文
1 W. Stallings. ISDN and Broadband ISDN with Frame Relay and ATM. AAAI-97 Spring Symposium: Cross-language Text and Speech Retrieval,2001,(2):581-592
2 陈讯编.宽带通信网信令及测试.北京:人民邮电出版社,2002:112-117
3 莫锦军.网络与ATM技术.北京:人民邮电出版社,2003:78-81
4 Craig Zacker.现代网络技术.王建华.北京:机械工业出版社,2002:21-26
5 Braden R., Clark D., Shenker S. Integrated Services in the Internet Architecture. Computer Networks and ISDN Systems,1994, (3):21-32
6 S. Blake, Yoram Bernet, James Binder. A Framework for Differentiated Services. http://www.gta.ufrj.br/diffserv/
7 D. Clark. Rethinking the Design of the Internet: End to End Arguments vs. the Brave New World. In Proc. AAAI/IAAI,1999,(12):78-81
8 S. Floyd. The Addition of Explicit Congestion Notification(ECN) to IP. http://www.icir.org/floyd/papers.html
9 Xipeng Xiao, Alan Hannan, Brook Bailey. Traffic Engineering with MPLS in Internet. IEEE Network,2000,(3):45-51
10 M. Hluchyjetal. Closed-Loop Rate-Based Traffic Management. ICDE Conference, 1994,(4):52-53
11 D. Hughes, P. Daley. Limitations of Credit Based Flow Control. Communications of the ACM,1999(3):24-32
12 Kawahara, P. Daley. More ABR Simulation Results. Applied Artificial Intelligence, 1994,(10):56-59
13 Duffield, Shirish Sathaye, K. Brinkerhoff. The Realities of Flow Control for ABR Service. Applied Artificial Intelligence,1999,(4):36-42
14 J. Jaffe. Bottleneck Flow Control. IEEE Trans. Comm,2003,l(29): 954-962
15 Elwalid, D. Chiu, W. Hawe. A Quantitative Measure of Fairness and Discrimination for Resource Allocation in Shared Systems. In Uournal of Information Retriecal, 1984,(12):87-91
Perros, Elsayed. The Art of Computer Systems Performance Analysis. New
16 York:John Wiley & Sons,1991:63-78
17 Gelenbe. Myths about Congestion Management in High Speed Networks. IEEE, 1992,(3):101-113
18 Gaiti. Congestion Control in Computer Networks: Issues and Trends. IEEE Network Magazine,1999,(3):24-30
19 Sriram. A Delay-Based Approach for Congestion Avoidance in Interconnected Heterogeneous Computer Networks. Computer Communications Review,1989,1(19) : 56-71
20 Bolla, K. K. Ramakrishnan, D. M. Chiu. Congestion Avoidance in Computer Networks with a Connectionless Network Layer. IEEE Network Magazine,1988,(3): 140-156
21 R. Jain, S. Kalyanaraman, R. Viswanathan. The OSU Scheme for Congestion Avoidance Using Explicit Rate Indication. Connection Science,2000,(5):12-16
22 Golestani. Comments on Rate-Based Proposal. Congestion Networks,1994,(5):21-26
23 Faber, A. Lin. Definition and Preliminary Simulation f a Rate-based Congestion Control Mechanism with Explicit Feedback of Bottleneck Rates. IEEE Transactions on Congestion Networks,1994,(2):23-28
24 R. Jain, S. Kalyanaraman, R. Viswanathan. A Sample Switch Algorithm. IEEE Transaction Congestion Networks,2000,(2):8-12
25 H. T. Kung. Adaptive Credit Allocation for Flow-Controlled VCs. IEEE Transactions on Congestion Networks,1994,(1):23-24
26 H. T. Kung. Flow Controlled Virtual Connections Proposal for ATM Traffic Management. IEEE Transactions on Congestion Networks,1994,(8):17-21
27 陈锡生.ATM交换技术.北京:人民邮电出版社,2000:65-68
28 P. Newman. Traffic Management for ATM Local Area Networks. IEEE Communications Magazine,2000,(2):44-50
29 张立明.人工神经网络的模型及其应用.上海:复旦大学出版社,1994:21-31
30 史忠科.神经网络控制理论.西安:西北工业大学出版社,2000:41-110
31 Saito. Congestion Avoidance in Computer Networks with a Connectionless Network Layer. Digital Equipment Corporation, 1987,(3):41-47
K. K. Ramakrishnan. Issues with Backward Explicit Congestion Notification based
32 Congestion Control. Connection Science,1993,(2):27-34
33 H T Kung, R Morris. Credit-Based Flow Control for ATM Networks. IEEE Network Magazine,1999:21-23
34 Cuiqing Yang, Alapati V. S. Reddy. A Taxonomy for Congestion Algorithm in Packet Switching Networks. IEEE Network Magazine,1995,(2):23-26
35 马丁.德.普瑞克.异步传递方式宽带ISDN技术.程时瑞,刘斌.北京:人民邮电出版社,1999:81-82
36 汪齐贤,冯玉珉.异步传输模式——ATM技术及应用.北京:中国铁道出版社,1998: 8-14
37 王喆.B-ISDN与ATM基础理论及应用.北京:中国铁道出版社,2001:106-181
38 陈锡生.ATM交换技术.北京:人民邮电出版社,2000:3-11
39 ATM Forum Traffic Management Specification. Version 4.0. 1999:45-48
40 S. Sathaye. Draft ATM Forum Traffic Management Specification Version 4.0. Computer Communications,1994,(4):41-46
41 L. Roberts. Rate-based Algorithm for Point to Multipoint ABR Service. Proceedings of 3rd ACM/IEEE MobiCom,Budapest,Hungary,2001:45-56
42 J. Scott. Link by Link Per VC Credit Based Flow Control. Communications of the ACM,1994,(5):24-26
43 K. K. Ramakrishnan, P. Newman. Credit Where Credit is Due. IEEE Transactions on Congestion Networks,1999,(3):41-47
44 R. Jain, S. Kalyanaraman, R. Goyal. ERICA+: Extensions to the ERICA Switch Algorithm. IEEE Transaction Congestion Networks,2000,(6):13-18
45 R.Jani, S.Fahmy, S.Kalyanaraman. ERICA Swith Algorithm:a Complete Description. ATM Forum Contribution,2001:24-28
46 R. Jain, S. Kalyanaraman, R. Viswanathan. The Transient Performance: EPRCA vs EPRCA+. Connection Science,1994,(2):6-10
47 K. Siu, H. Tzeng. Adaptive Proportional Rate Control for ABR Service in ATM Networks. University of California Irvine Technical,1994,(5):24-29
48 袁曾任.人工神经元网络及其应用.北京:清华大学出版社,1996:85-94
49 王耀南.智能控制系统.长沙:湖南大学出版社,1996:25-32
50 李学桥,马莉.神经网络工程应用.重庆:重庆大学出版社,1996:45-53
K W Fendick. An Approach to High Performance High-speed Data Networks. IEEE
51 Commu Magazine,1991:74-82
52 A.W.Barnhart. Baseline Performance Using PRCA Rate-control ATM Forum Contribution. J. Artificial Intelligence,1999,(6):27-32
53 L. Roberts. The Benefits of Rate-Based Flow Control for ABR Service. IEEE Transaction Congestion Networks,1994,3(4):52-61
54 F.M Chiussi, Y.T.Wang. An ABR Rate-based Congestion Control Algorithm for ATM Switches and Per-VC Quening. Proc.IEEE GLOBECOM,1999,2(11):771-778
55 R. Jain, S. Kalyanaraman, R. Viswanathan. The EPRCA Scheme. Connection Science,2000,(4):8-12
56 L.S.Brakmo, L.L.Peterson. TCP Vegas:End to End Congestion Avoidance on a Global Internet. IEEE J.Select.Areas Commun,1999,13(10):1465-1480
2 陈讯编.宽带通信网信令及测试.北京:人民邮电出版社,2002:112-117
3 莫锦军.网络与ATM技术.北京:人民邮电出版社,2003:78-81
4 Craig Zacker.现代网络技术.王建华.北京:机械工业出版社,2002:21-26
5 Braden R., Clark D., Shenker S. Integrated Services in the Internet Architecture. Computer Networks and ISDN Systems,1994, (3):21-32
6 S. Blake, Yoram Bernet, James Binder. A Framework for Differentiated Services. http://www.gta.ufrj.br/diffserv/
7 D. Clark. Rethinking the Design of the Internet: End to End Arguments vs. the Brave New World. In Proc. AAAI/IAAI,1999,(12):78-81
8 S. Floyd. The Addition of Explicit Congestion Notification(ECN) to IP. http://www.icir.org/floyd/papers.html
9 Xipeng Xiao, Alan Hannan, Brook Bailey. Traffic Engineering with MPLS in Internet. IEEE Network,2000,(3):45-51
10 M. Hluchyjetal. Closed-Loop Rate-Based Traffic Management. ICDE Conference, 1994,(4):52-53
11 D. Hughes, P. Daley. Limitations of Credit Based Flow Control. Communications of the ACM,1999(3):24-32
12 Kawahara, P. Daley. More ABR Simulation Results. Applied Artificial Intelligence, 1994,(10):56-59
13 Duffield, Shirish Sathaye, K. Brinkerhoff. The Realities of Flow Control for ABR Service. Applied Artificial Intelligence,1999,(4):36-42
14 J. Jaffe. Bottleneck Flow Control. IEEE Trans. Comm,2003,l(29): 954-962
15 Elwalid, D. Chiu, W. Hawe. A Quantitative Measure of Fairness and Discrimination for Resource Allocation in Shared Systems. In Uournal of Information Retriecal, 1984,(12):87-91
Perros, Elsayed. The Art of Computer Systems Performance Analysis. New
16 York:John Wiley & Sons,1991:63-78
17 Gelenbe. Myths about Congestion Management in High Speed Networks. IEEE, 1992,(3):101-113
18 Gaiti. Congestion Control in Computer Networks: Issues and Trends. IEEE Network Magazine,1999,(3):24-30
19 Sriram. A Delay-Based Approach for Congestion Avoidance in Interconnected Heterogeneous Computer Networks. Computer Communications Review,1989,1(19) : 56-71
20 Bolla, K. K. Ramakrishnan, D. M. Chiu. Congestion Avoidance in Computer Networks with a Connectionless Network Layer. IEEE Network Magazine,1988,(3): 140-156
21 R. Jain, S. Kalyanaraman, R. Viswanathan. The OSU Scheme for Congestion Avoidance Using Explicit Rate Indication. Connection Science,2000,(5):12-16
22 Golestani. Comments on Rate-Based Proposal. Congestion Networks,1994,(5):21-26
23 Faber, A. Lin. Definition and Preliminary Simulation f a Rate-based Congestion Control Mechanism with Explicit Feedback of Bottleneck Rates. IEEE Transactions on Congestion Networks,1994,(2):23-28
24 R. Jain, S. Kalyanaraman, R. Viswanathan. A Sample Switch Algorithm. IEEE Transaction Congestion Networks,2000,(2):8-12
25 H. T. Kung. Adaptive Credit Allocation for Flow-Controlled VCs. IEEE Transactions on Congestion Networks,1994,(1):23-24
26 H. T. Kung. Flow Controlled Virtual Connections Proposal for ATM Traffic Management. IEEE Transactions on Congestion Networks,1994,(8):17-21
27 陈锡生.ATM交换技术.北京:人民邮电出版社,2000:65-68
28 P. Newman. Traffic Management for ATM Local Area Networks. IEEE Communications Magazine,2000,(2):44-50
29 张立明.人工神经网络的模型及其应用.上海:复旦大学出版社,1994:21-31
30 史忠科.神经网络控制理论.西安:西北工业大学出版社,2000:41-110
31 Saito. Congestion Avoidance in Computer Networks with a Connectionless Network Layer. Digital Equipment Corporation, 1987,(3):41-47
K. K. Ramakrishnan. Issues with Backward Explicit Congestion Notification based
32 Congestion Control. Connection Science,1993,(2):27-34
33 H T Kung, R Morris. Credit-Based Flow Control for ATM Networks. IEEE Network Magazine,1999:21-23
34 Cuiqing Yang, Alapati V. S. Reddy. A Taxonomy for Congestion Algorithm in Packet Switching Networks. IEEE Network Magazine,1995,(2):23-26
35 马丁.德.普瑞克.异步传递方式宽带ISDN技术.程时瑞,刘斌.北京:人民邮电出版社,1999:81-82
36 汪齐贤,冯玉珉.异步传输模式——ATM技术及应用.北京:中国铁道出版社,1998: 8-14
37 王喆.B-ISDN与ATM基础理论及应用.北京:中国铁道出版社,2001:106-181
38 陈锡生.ATM交换技术.北京:人民邮电出版社,2000:3-11
39 ATM Forum Traffic Management Specification. Version 4.0. 1999:45-48
40 S. Sathaye. Draft ATM Forum Traffic Management Specification Version 4.0. Computer Communications,1994,(4):41-46
41 L. Roberts. Rate-based Algorithm for Point to Multipoint ABR Service. Proceedings of 3rd ACM/IEEE MobiCom,Budapest,Hungary,2001:45-56
42 J. Scott. Link by Link Per VC Credit Based Flow Control. Communications of the ACM,1994,(5):24-26
43 K. K. Ramakrishnan, P. Newman. Credit Where Credit is Due. IEEE Transactions on Congestion Networks,1999,(3):41-47
44 R. Jain, S. Kalyanaraman, R. Goyal. ERICA+: Extensions to the ERICA Switch Algorithm. IEEE Transaction Congestion Networks,2000,(6):13-18
45 R.Jani, S.Fahmy, S.Kalyanaraman. ERICA Swith Algorithm:a Complete Description. ATM Forum Contribution,2001:24-28
46 R. Jain, S. Kalyanaraman, R. Viswanathan. The Transient Performance: EPRCA vs EPRCA+. Connection Science,1994,(2):6-10
47 K. Siu, H. Tzeng. Adaptive Proportional Rate Control for ABR Service in ATM Networks. University of California Irvine Technical,1994,(5):24-29
48 袁曾任.人工神经元网络及其应用.北京:清华大学出版社,1996:85-94
49 王耀南.智能控制系统.长沙:湖南大学出版社,1996:25-32
50 李学桥,马莉.神经网络工程应用.重庆:重庆大学出版社,1996:45-53
K W Fendick. An Approach to High Performance High-speed Data Networks. IEEE
51 Commu Magazine,1991:74-82
52 A.W.Barnhart. Baseline Performance Using PRCA Rate-control ATM Forum Contribution. J. Artificial Intelligence,1999,(6):27-32
53 L. Roberts. The Benefits of Rate-Based Flow Control for ABR Service. IEEE Transaction Congestion Networks,1994,3(4):52-61
54 F.M Chiussi, Y.T.Wang. An ABR Rate-based Congestion Control Algorithm for ATM Switches and Per-VC Quening. Proc.IEEE GLOBECOM,1999,2(11):771-778
55 R. Jain, S. Kalyanaraman, R. Viswanathan. The EPRCA Scheme. Connection Science,2000,(4):8-12
56 L.S.Brakmo, L.L.Peterson. TCP Vegas:End to End Congestion Avoidance on a Global Internet. IEEE J.Select.Areas Commun,1999,13(10):1465-1480