PhD graduated
Team : NPA
Departure date : 09/30/2014

Supervision : Serge FDIDA

Co-supervision : FRIEDMAN Timur

IP-level Internet topology dynamics capture

Large-scale distributed network route tracing systems obtain the IP-level internet topology by orchestrating Traceroute-like tool on distributed sources that capture the paths to several destinations in the network. This information is important to network operators to better monitor the state of their network, react to failures and detect anomalies while researchers use it to model and understand the behavior of the underlying network topology. However, existing approaches to measuring the public IPv4 network space often require one or more days to obtain a full graph, which is too slow to capture much of the network's dynamics.Thus, new network route tracing system algorithms are highly expected to reach the requirements of scalability, measurement speed, and dynamics capture accuracy.
This thesis proposes a fundamental improvement to distributed IP-level Internet topology measurement systems. Such systems have focused on conducting full end-to-end route traces and as a result take a considerable time to obtain a graph of the network. Here, we propose an approach focused directly on obtaining the graph, which is, as a result, much faster (presuming the principal goal is to obtain this graph, rather than full routes). The main benefit of our approach is that we can get a much better view of the dynamics of the network at the IP-level. In addition to IP-level measurement, this thesis develops a novel approach to network topology services by designing a federated monitoring infrastructure that exposes a wide range of measurement metrics for testbed users.

Defence : 11/27/2013 - 14h00 - Site Jussieu 25-26/105

Jury members :

Damien MAGONI, LaBRI [Rapporteur]
Javier ARACIL, UAM [Rapporteur]
Dimitri PAPADIMITRIOU, Alcatel-Lucent

2009-2016 Publications