Testbeds and Research Infrastructure: Development of Networks and Communities. 9th International ICST Conference, TridentCom 2014, Guangzhou, China, May 5-7, 2014, Revised Selected Papers

Research Article

An Early Traffic Sampling Algorithm

Download58 downloads
  • @INPROCEEDINGS{10.1007/978-3-319-13326-3_41,
        author={Hou Ying and Huang Hai and Chen Dan and Wang ShengNan and Li Peng},
        title={An Early Traffic Sampling Algorithm},
        proceedings={Testbeds and Research Infrastructure: Development of Networks and Communities. 9th International ICST Conference, TridentCom 2014, Guangzhou, China, May 5-7, 2014, Revised Selected Papers},
        proceedings_a={TRIDENTCOM},
        year={2014},
        month={11},
        keywords={Traffic classification Bloom Filter Early Traffic Sampling},
        doi={10.1007/978-3-319-13326-3_41}
    }
    
  • Hou Ying
    Huang Hai
    Chen Dan
    Wang ShengNan
    Li Peng
    Year: 2014
    An Early Traffic Sampling Algorithm
    TRIDENTCOM
    Springer
    DOI: 10.1007/978-3-319-13326-3_41
Hou Ying1,*, Huang Hai1,*, Chen Dan1,*, Wang ShengNan1, Li Peng1
  • 1: National Digital Switching System Engineering & Techological R&D center
*Contact email: ndschy@139.com, hh@mail.ndsc.com.cn, cd@mail.ndsc.com.cn

Abstract

The first several packets of a flow play key role in the on-line traffic managements. Early traffic sampling, extracting the first several packets of every flow, is raised. This paper proposes a structure named CTBF, combination of counting Bloom Filter and time Bloom Filter. Based on it, the algorithm is designed to realize automatically removing the space occupied by the timeout flow. The analyses and experiments demonstrate that the sampling accuracy of CTBF is better than that of LRU and Fixed-T algorithm in the same space.