Poster Abstract: A flexible infinite HMM model for accurate characterization and segmentation of RTT timeseries

Abstract : The study of round-trip time (RTT) measurements on the Internet is of particular importance for improving real-time applications, enforcing QoS with traffic engineering, or detecting unexpected network conditions. On large timescales, from 1 hour to several days, RTT measurements exhibit characteristic patterns due to inter and intra-AS routing changes and traffic engineering, in addition to link congestion. We propose the use of a nonparametric Bayesian method to fully estimate HMM parameters from delay observations, including the number of states. We validate the model through three applications: the clustering of RIPE Atlas measurements, the detection of significant delay changes, and the reduction of the monitoring cost in routing overlays using Markov decision processes.
Document type :
Conference papers
Complete list of metadatas

Cited literature [4 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-02300968
Contributor : Maxime Mouchet <>
Submitted on : Monday, September 30, 2019 - 9:50:59 AM
Last modification on : Thursday, October 17, 2019 - 12:36:56 PM

File

INFOCOM_2019___Abstract.pdf
Files produced by the author(s)

Identifiers

Citation

Maxime Mouchet, Sandrine Vaton, Thierry Chonavel. Poster Abstract: A flexible infinite HMM model for accurate characterization and segmentation of RTT timeseries. IEEE INFOCOM 2019 - IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS), Apr 2019, Paris, France. pp.1055-1056, ⟨10.1109/INFCOMW.2019.8845296⟩. ⟨hal-02300968⟩

Share

Metrics

Record views

56

Files downloads

66