[1]刘卫江,白磊,杨望.基于TCBF算法的TCP流长度分布统计[J].东南大学学报(自然科学版),2007,37(5):787-792.[doi:10.3969/j.issn.1001-0505.2007.05.010]
 Liu Weijiang,Bai Lei,Yang Wang.Statistics of distribution of TCP flow length based on TCBF algorithm[J].Journal of Southeast University (Natural Science Edition),2007,37(5):787-792.[doi:10.3969/j.issn.1001-0505.2007.05.010]
点击复制

基于TCBF算法的TCP流长度分布统计()
分享到:

《东南大学学报(自然科学版)》[ISSN:1001-0505/CN:32-1178/N]

卷:
37
期数:
2007年第5期
页码:
787-792
栏目:
计算机科学与工程
出版日期:
2007-09-20

文章信息/Info

Title:
Statistics of distribution of TCP flow length based on TCBF algorithm
作者:
刘卫江13 白磊2 杨望3
1 大连海事大学计算机科学与技术学院, 大连 116026; 2 华北科技学院计算机系, 廊坊 101601; 3 东南大学计算机科学与工程学院,南京 210096
Author(s):
Liu Weijiang13 Bai Lei2 Yang Wang3
1 School of Computer Science and Technology, Dalian Maritime University, Dalian 116026, China
2 Deparment of Computer Science and Technology, North China Institute of Science and Technology, Langfang 101601, China
3
关键词:
报文抽样 报文过滤 TCP流 重尾分布
Keywords:
packet sampling packet filtering TCP flow heavy-tailed distribution
分类号:
TP393
DOI:
10.3969/j.issn.1001-0505.2007.05.010
摘要:
为提高流测量系统的运行效率,减少其所需资源的消耗,在分析了TCP连接的释放以发送FIN报文或RST报文作为结束标志的基础上,提出一种新的报文过滤算法TCBF(time-out counting bloom filter),用于统计TCP流长度分布.通过对有结束标志的流、没有结束标志的中长流和没有结束标志的中短流分别采用不同的统计处理方式,增加了网络测量性能,提高了测量系统的资源利用率.该算法不需要收集所有报文,而是只需要抽样少量报文即可对流长度做出统计.理论分析和仿真结果表明,TCBF算法在使用较少的存储空间的条件下,可以及时准确地对TCP流长度分布做出统计,满足实际测量需要.
Abstract:
In order to improve the performance and reduce the resource usage of flow-based measurement systems, a novel packet filtering algorithm for generating the statistics of the distribution of transmission control protocol(TCP)flow length, time-out counting bloom filter(TCBF), is presented based on the characteristics analysis of that sending a FIN or RST segment is the termination flag of closing a TCP connection in the Internet. This algorithm can improve the performance of network measurement and the efficiency of the resource usage of measurement system by handling flows with termination flag, large flows without termination and middle-short flows without termination with different strategy. Instead of collecting all packets this algorithm may make statistics of flow length with only a few packets. The theoretical analysis and the simulation results indicate that the TCBF algorithm can generate the statistics of the distribution of TCP flow length timely and accurately under the condition of using less storage space, and satisfy the need of practical measurement.

参考文献/References:

[1] Gong J,Cheng G.Distributed sampling measurement model in a large-scale high-speed IP networks[J]. Journal of Southeast University:English Edition,2002,18(1):40-45.
[2] Mori Tatsuya,Uchida Masato,Kawahara Ryoichi,et al.Identifying elephant flows through periodically sampled packets[C] //Proc of ACM SIGCOMM/IMC’04.Taormina,Italy,2004:115-120.
[3] Duffield N,Lund C,Thorup M.Charging from sampled network usage[C] //Proc ACM SIGCOMM Internet Measurement Workshop.San Francisco,CA,USA,2001:245-256.
[4] Duffield N,Lund C,Thorup M.Properties and prediction of flow statistics from sampled packet streams[C] //Proc ACM SIGCOMM Internet Measurement Workshop.Marseille,France,2002:159-171.
[5] Feldmann A,Caceres R,Douglis F,et al.Performance of web proxy caching in heterogeneous bandwidth environments[C] //Proc IEEE INFOCOM.New York,1999:107-116.
[6] Feldmann A,Rexford J,Caceres R.Efficient policies for carrying web traffic over flow-switched networks [J].Proc IEEE/ACM Transactions on Networking,1998,6(6):673-685.
[7] Kompella R,Singh S,Varghese G.On scalable attack detection in the network[C] //Proc ACM SIGCOMM Internet Measurement Conference.Taormina,Italy,2004:187-200.
[8] Duffield N,Lund C,Thorup M.Estimating flow distributions from sampled flow statistics[J]. IEEE/ACM Transactions on Networking,2005,13(5):933-945.
[9] Andrei B,Michael M.Network applications of bloom filters:a survey[J]. Internet Math,2003(4):485-509.
[10] Estan C,Varghese G.New directions in traffic measurement and accounting[C] //Proc ACM SIGCOMM.San Francisco,CA,USA,2001:75-80.
[11] Kumar A,Xu J,Li L,et al.Space-code bloom filter for efficient traffic flow measurement[C] //Proc ACM SIGCOMM/IMC.Miami Beach,FL,USA,2003:167-172.
[12] Kumar A,Sung M,Xu J,et al.Data streaming algorithms for efficient and accurate estimation of flow size distribution[C] //Proc ACM SIGMETRICS.New York,2004:177-188.
[13] Brownlee N,Mills C,Ruth G.RFC 2722 traffic flow measurement:architecture[S].IETF,1999.
[14] 周明中,龚俭,丁伟.网络流超时策略研究[J].通信学报,2005,26(4):88-93.
  Zhou Mingzhong,Gong Jian,Ding Wei.Study of network flow timeout strategy[J].Journal on Communications,2005,26(4):88-93.(in Chinese)
[15] Kong Shijin,He Tao,Shao Xiaoxin,et al.Time-out bloom filter:a new sampling method for recording more flows[C] //Lecture Notes in Computer Science 3961,Proc ICOIN 2006.Sendai,Japan,2006:590-599.

备注/Memo

备注/Memo:
基金项目: 国家重点基础研究发展计划(973计划)资助项目(2003CB314804)、教育部科学技术研究重点资助项目(105084)、 江苏省网络与信息安全重点实验室资助项目(BM2003201)、江苏省博士后科研资助计划资助项目.
作者简介: 刘卫江(1969—),男,博士,教授,wjliu@njnet.edu.cn.
更新日期/Last Update: 2007-09-20