[1] Gong Jian**, Cheng Guang,. Distributed Sampling Measurement Model ina Large-Scale High-Speed IP Networks* [J]. Journal of Southeast University (English Edition), 2002, 18 (1): 40-45. [doi:10.3969/j.issn.1003-7985.2002.01.008]
Copy
Distributed Sampling Measurement Model ina Large-Scale High-Speed IP Networks*(
)
大规模高速IP网络分布式抽样测量模型
)Journal of Southeast University (English Edition)[ISSN:1003-7985/CN:32-1325/N]
- Volumn:
- 18
- Issue:
- 2002 1
- Page:
- 40-45
- Research Field:
- Computer Science and Engineering
- Publishing date:
- 2002-03-30
Info
- Title:
- Distributed Sampling Measurement Model ina Large-Scale High-Speed IP Networks*
- 大规模高速IP网络分布式抽样测量模型
- Author(s):
- Gong Jian**; Cheng Guang
- Department of Computer Science and Engineering, Southeast University, Nanjing 210096, China
- PACS:
- TP393.07
- DOI:
- 10.3969/j.issn.1003-7985.2002.01.008
- Abstract:
- The distributed passive measurement is an important technology for network behavior research. To achieve a consistent measurement, the same packets should be sampled at distributed measurement points. And in order to estimate the character of traffic statistics, the traffic sample should be random in statistics. A distributed sampling mask measurement model is introduced to tackle the difficulty of measuring the full trace of high-speed networks. The key point of the model is to choose some bits that are suitable to be sampling mask. In the paper, the bit entropy and bit flow entropy of IP packet headers in CERNET backbone are analyzed, and we find that the 16 bits of identification field in IP packet header are fit to the matching field of sampling mask. Measurement traffic also can be used to analyze the statistical character of measurement sample and the randomicity of the model. At the same time the experiment results indicate that the model has a good sampling performance.
- 分布式被动测量是研究网络行为的一个重要手段.为了获得分布式协同处理流量信息, 要求分布的测量点能抽取同样报文;为了能估计流量总体统计属性, 抽样样本需要具有统计随机性.为此, 文章提出分布式掩码抽样测量模型处理高速网络流量, 其核心是确定合适的抽样掩码匹配位串.对CERNET主干网络流量IP报头各字段的位熵和位流熵进行分析, 结果表明标识字段16比特适合于抽样掩码匹配字段.使用测量数据分析基于标识字段抽样模型的随机性和抽样样本的统计属性, 实验进一步验证了所提出的模型具有良好的抽样性.
References:
[1] Huffaker B, Fomenkov M, Moore D, et al. Measurements of the internet topology in Asia-pacific region[EB/OL]. http://www.caida.org/outreach/papers/asia-paper/, 2000.
[2] Graham I D, Donnelly S F, Martin S, et al. Nonintrusive and accurate measurement of unidirectional delay and delay variation on the internet[A]. In: Proc. INET’98[C]. July 1998.
[3] Zseby T, Zander S, Carle G. Evaluation of build blocks for passive one-way-delay measurements, PAM2001, In Amsterdam USA, 2001.10.
[4] Duffield N, Grossglauser M. Trajectory sampling for direct traffic observation[A]. In: Proceedings of ACM SIGCOMM 2000[C]. Stockholm, Sweden, August 28-September 1, 2000.
[5] Cozzani I, Giordano S. A passive test and measurement system: traffic sampling for QoS evaluation[A]. In: Proc. of IEEE Globecom’98 Sydney[C]. Australia, November 1998.
[6] Claffy K, Polyzos G, Braun H. Application of sampling methodologies to network traffic characterization[A]. In: Proceedings of ACM SIGCOMM’93[C]. May 1993.
[7] Drobisz J, Christensen K J. Adaptive sampling methods to determine network traffic statistics including the hurst parameter[A]. 23rd Annual Conference on Local Computer Networks[C]. October 11-14, 1998.
[8] Paxson V, Almes G, Mahdavi J, Mathis M. Framework for IP performance metrics, IETF RFC 2330, 1998.
[9] Jin Zhenyu. Information theory[M]. Beijing: Beijing University of Science and Technology Press, 1991.11-47.(in Chinese)
[10] Reynolds J, Postel J. Assigned numbers, IETF RFC1700, October 1994.
[11] Tang Xiangneng, Dai Jianhua. Mathematics Statistics[M]. Beijing: Mechanism Technology Press, 1994.140-151.(in Chinese)
Memo
- Memo:
- * The project supported by the National Natural Science Foundation of China(90104031), and 863 program of China(2001AA112060).
** Born in 1957, male, professor.