关于传输虚拟化中数据分组乱序问题的研究

doi:10.3969／j.issn.1000-0801.2012.10.014

摘要/Abstract

摘要：

对传输虚拟化相关问题及传输虚拟化中多路径并发传输的数据分组乱序问题进行了介绍，从静态的路径选择机制以及动态的数据分组调度模型方面分析了造成接收端数据分组乱序的主要原因，提出了基于路径往返时间的路径选择机制和基于权重轮询调度算法、时延均衡技术的数据分组调度模型。理论分析和仿真实验的结果表明，所提方案能够明显地缓解接收端数据分组的乱序问题，有效地降低了接收端缓存区容量的需求。

关键词: 传输虚拟化, 多路径并发传输, 多路径选择机制, 数据分组调度模型, 数据分组乱序

Abstract:

This paper introduces the problem of transport virtualization and the packet reordering on concurrent multipath transmissions for transport virtualization, analyzes the main reasons for the packet reordering from the point of static path selection mechanism, as well as dynamic packet scheduling model. A new path selection method based on round trip time（RTT） and a packet scheduling model based on weighted round-robin scheduling algorithm and delay equalizer technology are proposed. Theoretical analysis and simulation experiment results show that the scheme can significantly ease the packets reordering and effectively reduces the require of receiver buffer capacity.

Key words: transport virtualization, concurrent multipath transmission, multipath selection method, packet scheduling model, packet reordering

葛杨,徐名海,迟欢. 关于传输虚拟化中数据分组乱序问题的研究[J]. 电信科学, 2012, 28(10): 80-87.

Yang Ge,Minghai Xu,Huan Chi. Research on the Packet Reordering in Transport Virtualization[J]. Telecommunications Science, 2012, 28(10): 80-87.

图/表 10

图1

图2

图3

图4

图5

表1

图6

表2

图7

表3

参考文献 12

1	Hoffmann M , Staufer M . Network virtualization for future mobile networks: general architecture and applications. 2011 IEEE International Conference on Communications Workshops（ICC）, ACM, Kyoto, Japan, 2011
2	Tutschku K , Zinner T , Nakao A , et al. Network virtualization:implementation steps towards the future internet. Proceedings of the Workshop on Overlay and Network Virtualization at KiVS 2009, Kassel, Germany, 2009
3	Zinner T , Klein D , Tutschku K , et al. Performance of concurrent multipath transmissions-measurements and model validation. Proceedings of the 7th EURO-NGI Conference, Kaiserslautern, Germany, 2011
4	Tutschku K , Zinner T , Nakao A , et al. Re-sequencing buffer occupancy of a concurrent multipath transmission mechanism for transport system virtualization. Proceedings of the 16 ITG/GI -Fachtagung Kommunikation in Verteilten Systemen 2009（KiVS 2009）, Kassel, Germany, Mar 2009
5	Zinner T , Tutschku K , Nakao A . Using concurrent multipath transmission for transport virtualization: analyzing path selection. Proceedings of the 22nd International Telegraphic Congress （ITC）, Amsterdam, Netherlands, 2010
6	Lane J , Nakao A . Sora: a shared overlay routing architecture. Proceedings of the 2nd International Workshop on Real Overlays and Distributed Systems（ROADS）, War-Saw, Poland, 2007
7	Gummadi K , Madhyastha H , Gribble S , et al. Improving the reliability of internet paths with one-hop source routing. Proceedings of OSDI'04, San Francisco, CA, USA, 2004
8	Lane J , Nakao A . Best-effort network layer packet reordering in support of multi-path overlay packet dispersion. Proceedings of Globecom08, New Orleans, LA, 2008
9	Kaspar D , Evensen K , Hansen A F , et al. An analysis of the heterogeneity and IP packet reordering over multiple wireless networks. IEEE Symposium on Computers and Communications （ISCC）, USA, 2009
10	Piratla N M , Jayasumana A P . Metrics for packet reordering-a comparative analysis. International Journal of Communication Systems, 2008, 21（1）:99～113
11	Morton A , Ciavattone L , Ramachandran G , et al. Packet Reordering Metrics. IETF RFC4737, 2006
12	Jayasumana A , Piratla N , Banka T , et al. Improved Packet Reordering Metrics. IETF RFC5236, 2008

数据分组序列号	1	2	3	4	5	6	7	8	9	10	11
发送矢量	0	0	1	0	0	1	0	1	0	0	1
到达NV-boxes的时间（ms）	0	2	4	6	8	10	12	14	16	18	20
等待时间偏差T_i（ms）	0	0	20	0	0	6	0	4	0	0	6
总的等待时间T_toti（ms）	0	0	20	0	0	20	0	20	0	0	20
离开NV-boxes的时间（ms）	0	2	24	6	8	30	12	34	16	18	40
到达对端的时间（ms）	30	32	34	36	38	40	42	44	46	48	50

带宽比率	1:11	2:10	3:9	4:8	5:7	6:6
平均吞吐量（Mbit/s）	5.99	5.99	5.99	5.99	5.99	5.99
数据分组乱序率	1.56%	1.98%	1.68%	1.29%	2.03%	0.09%
99%缓存区平均占用率（数据分组数）	1	1	1	1	1	0
分组丢失数	0.10	0.12	0.11	0.05	0.04	0.01

	round-robin调度模型	改进的调度模型
平均吞吐量（Mbit/s）	4.49	5.86
99%缓存区占用率（数据分组数）	26	9
分组丢失数	25%	2.2%