TU Darmstadt / ULB / TUbiblio

Capturing Connectivity Graphs of a Large-Scale P2P Overlay Network

Salah, Hani ; Strufe, Thorsten (2013)
Capturing Connectivity Graphs of a Large-Scale P2P Overlay Network.
Konferenzveröffentlichung, Bibliographie

Kurzbeschreibung (Abstract)

Measuring accurate graph snapshots of peer-to-peer (P2P) overlay networks is essential to understand these systems. Furthermore, the captured graph snapshots can be used, among other important applications, as traces for simulation studies, to validate existing simulation models, to design and implement targeted attacks, or to detect anomalies. Motivated by the importance of the applications above as well as the popularity of several Kademlia-like networks, we present a new crawler that is able, for the first time, to capture snapshots of the connectivity graph of the entire KAD network. The crawler’s design is generic and can measure other Kademlia-like networks, and it is adaptable to other P2P networks. Our results show that our crawler is fast and captures accurate graph snapshots. Its design enables it to outperform prior KAD crawlers in terms of the time and the number of crawling messages that are required to download nodes’ routing tables. The crawls that we conducted at different times between April 2012 and February 2013 show that KAD is still widely-used in terms of total observed users. However, when compared to the results of prior studies, we report a significant drop in the number of simultaneous online users.

Typ des Eintrags: Konferenzveröffentlichung
Erschienen: 2013
Autor(en): Salah, Hani ; Strufe, Thorsten
Art des Eintrags: Bibliographie
Titel: Capturing Connectivity Graphs of a Large-Scale P2P Overlay Network
Sprache: Deutsch
Publikationsjahr: 2013
Buchtitel: 5th IEEE ICDCS International Workshop on Hot Topics in Peer-to-Peer Computing and Online Social (HotPOST)
Kurzbeschreibung (Abstract):

Measuring accurate graph snapshots of peer-to-peer (P2P) overlay networks is essential to understand these systems. Furthermore, the captured graph snapshots can be used, among other important applications, as traces for simulation studies, to validate existing simulation models, to design and implement targeted attacks, or to detect anomalies. Motivated by the importance of the applications above as well as the popularity of several Kademlia-like networks, we present a new crawler that is able, for the first time, to capture snapshots of the connectivity graph of the entire KAD network. The crawler’s design is generic and can measure other Kademlia-like networks, and it is adaptable to other P2P networks. Our results show that our crawler is fast and captures accurate graph snapshots. Its design enables it to outperform prior KAD crawlers in terms of the time and the number of crawling messages that are required to download nodes’ routing tables. The crawls that we conducted at different times between April 2012 and February 2013 show that KAD is still widely-used in terms of total observed users. However, when compared to the results of prior studies, we report a significant drop in the number of simultaneous online users.

Freie Schlagworte: Measurement, Crawler, KAD, Connectivity Graph
ID-Nummer: TUD-CS-2013-0072
Fachbereich(e)/-gebiet(e): 20 Fachbereich Informatik > Peer-to-Peer Netzwerke
20 Fachbereich Informatik
Hinterlegungsdatum: 27 Jul 2016 16:32
Letzte Änderung: 17 Mai 2018 21:50
PPN:
Export:
Suche nach Titel in: TUfind oder in Google
Frage zum Eintrag Frage zum Eintrag

Optionen (nur für Redakteure)
Redaktionelle Details anzeigen Redaktionelle Details anzeigen