Large-Scale Characterization and Segmentation of Internet Path Delays With Infinite HMMs
Round-Trip Times are one of the most commonly collected performance metrics in computer networks. Measurement platforms such as RIPE Atlas provide researchers and network operators with an unprecedented amount of historical Internet delay measurements. It would be very useful to process these measur...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/8964300/ |
| _version_ | 1818853393006854144 |
|---|---|
| author | Maxime Mouchet Sandrine Vaton Thierry Chonavel Emile Aben Jasper Den Hertog |
| author_facet | Maxime Mouchet Sandrine Vaton Thierry Chonavel Emile Aben Jasper Den Hertog |
| author_sort | Maxime Mouchet |
| collection | DOAJ |
| description | Round-Trip Times are one of the most commonly collected performance metrics in computer networks. Measurement platforms such as RIPE Atlas provide researchers and network operators with an unprecedented amount of historical Internet delay measurements. It would be very useful to process these measurements automatically (statistical characterization of path performance, change detection, recognition of recurring patterns, etc.). Humans are quite good at finding patterns in network measurements, but it can be difficult to automate this and enable many time series to be processed at the same time. In this article we introduce a new model, the HDP-HMM or infinite hidden Markov model, whose performance in trace segmentation is very close to human cognition. We demonstrate, on a labeled dataset and on RIPE Atlas and CAIDA MANIC data, that this model represents measured RTT time series much more accurately than classical mixture or hidden Markov models. This method is implemented in RIPE Atlas and we introduce the publicly accessible Web API. An interactive notebook for exploring the API is available on GitHub. |
| first_indexed | 2024-12-19T07:36:06Z |
| format | Article |
| id | doaj.art-1fe0d6296914471eb3d9f9921014fd1b |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-19T07:36:06Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-1fe0d6296914471eb3d9f9921014fd1b2022-12-21T20:30:35ZengIEEEIEEE Access2169-35362020-01-018167711678410.1109/ACCESS.2020.29683808964300Large-Scale Characterization and Segmentation of Internet Path Delays With Infinite HMMsMaxime Mouchet0https://orcid.org/0000-0002-7302-4818Sandrine Vaton1https://orcid.org/0000-0001-8940-6004Thierry Chonavel2https://orcid.org/0000-0003-3406-0426Emile Aben3https://orcid.org/0000-0002-8275-5460Jasper Den Hertog4https://orcid.org/0000-0001-6884-4937Lab-STICC, IMT Atlantique, Plouzané, FranceLab-STICC, IMT Atlantique, Plouzané, FranceLab-STICC, IMT Atlantique, Plouzané, FranceRIPE NCC, Amsterdam, The NetherlandsRIPE NCC, Amsterdam, The NetherlandsRound-Trip Times are one of the most commonly collected performance metrics in computer networks. Measurement platforms such as RIPE Atlas provide researchers and network operators with an unprecedented amount of historical Internet delay measurements. It would be very useful to process these measurements automatically (statistical characterization of path performance, change detection, recognition of recurring patterns, etc.). Humans are quite good at finding patterns in network measurements, but it can be difficult to automate this and enable many time series to be processed at the same time. In this article we introduce a new model, the HDP-HMM or infinite hidden Markov model, whose performance in trace segmentation is very close to human cognition. We demonstrate, on a labeled dataset and on RIPE Atlas and CAIDA MANIC data, that this model represents measured RTT time series much more accurately than classical mixture or hidden Markov models. This method is implemented in RIPE Atlas and we introduce the publicly accessible Web API. An interactive notebook for exploring the API is available on GitHub.https://ieeexplore.ieee.org/document/8964300/Round-trip timesRIPE Atlashidden Markov modelsnonparametric Bayesian modelsanomaly detectiontime series clustering |
| spellingShingle | Maxime Mouchet Sandrine Vaton Thierry Chonavel Emile Aben Jasper Den Hertog Large-Scale Characterization and Segmentation of Internet Path Delays With Infinite HMMs IEEE Access Round-trip times RIPE Atlas hidden Markov models nonparametric Bayesian models anomaly detection time series clustering |
| title | Large-Scale Characterization and Segmentation of Internet Path Delays With Infinite HMMs |
| title_full | Large-Scale Characterization and Segmentation of Internet Path Delays With Infinite HMMs |
| title_fullStr | Large-Scale Characterization and Segmentation of Internet Path Delays With Infinite HMMs |
| title_full_unstemmed | Large-Scale Characterization and Segmentation of Internet Path Delays With Infinite HMMs |
| title_short | Large-Scale Characterization and Segmentation of Internet Path Delays With Infinite HMMs |
| title_sort | large scale characterization and segmentation of internet path delays with infinite hmms |
| topic | Round-trip times RIPE Atlas hidden Markov models nonparametric Bayesian models anomaly detection time series clustering |
| url | https://ieeexplore.ieee.org/document/8964300/ |
| work_keys_str_mv | AT maximemouchet largescalecharacterizationandsegmentationofinternetpathdelayswithinfinitehmms AT sandrinevaton largescalecharacterizationandsegmentationofinternetpathdelayswithinfinitehmms AT thierrychonavel largescalecharacterizationandsegmentationofinternetpathdelayswithinfinitehmms AT emileaben largescalecharacterizationandsegmentationofinternetpathdelayswithinfinitehmms AT jasperdenhertog largescalecharacterizationandsegmentationofinternetpathdelayswithinfinitehmms |