Estratégias de Sondagem para Remapeamento Eficiente de Eventos de Roteamento na Internet

Path changes caused by events such as traffic engineering, changing traffic exchange partnerships, or link failures impact multiple paths on the Internet. Topological monitoring platforms perform periodic measurements using traceroute for a large number of destinations. This approach, however, is inadequate to precisely identify the extent of the impact of routing events. For example, a link failure can be restored before all routes are measured. In this paper we present measurement strategies that minimize the cost of probing to identify paths impacted by a routing event. Our results show that it is possible to identify the set of paths impacted by an event efficiently. Our results further indicate that when integrated into a state-of-the-art pathway tracking system, our strategies more than double the number of changes detected.

[1]  Ítalo S. Cunha,et al.  DTRACK: A System to Predict and Track Internet Path Changes , 2014, IEEE/ACM Transactions on Networking.

[2]  Albert G. Greenberg,et al.  Detection and Localization of Network Black Holes , 2007, IEEE INFOCOM 2007 - 26th IEEE International Conference on Computer Communications.

[3]  Ítalo S. Cunha,et al.  Efficient Remapping of Internet Routing Events , 2016, SIGCOMM.

[4]  Ítalo S. Cunha,et al.  LIFEGUARD: practical repair of persistent route failures , 2012, SIGCOMM '12.

[5]  Ítalo S. Cunha,et al.  Sibyl: A Practical Internet Route Oracle , 2016, NSDI.

[6]  Guillaume Urvoy-Keller,et al.  Characterizing ICMP rate limitation on routers , 2015, 2015 IEEE International Conference on Communications (ICC).

[7]  Jonathan M. Smith,et al.  Pushing the Boundaries with bdrmapIT: Mapping Router Ownership at Internet Scale , 2018, Internet Measurement Conference.

[8]  Ítalo S. Cunha,et al.  A Characterization of Load Balancing on the IPv6 Internet , 2017, PAM.

[9]  Dmitri Krioukov,et al.  Internet Mapping: From Art to Science , 2009, 2009 Cybersecurity Applications & Technology Conference for Homeland Security.

[10]  Matthew J. Luckie,et al.  Scamper: a scalable and extensible packet prober for active measurement of the internet , 2010, IMC '10.

[11]  Vasileios Giotsas,et al.  Periscope: Unifying Looking Glass Querying , 2016, PAM.

[12]  Vern Paxson,et al.  Packetlab: a universal measurement endpoint interface , 2017, Internet Measurement Conference.

[13]  Vasileios Giotsas,et al.  Mapping peering interconnections to a facility , 2015, CoNEXT.

[14]  Ítalo S. Cunha,et al.  Measuring and Characterizing End-to-End Route Dynamics in the Presence of Load Balancing , 2011, PAM.

[15]  Brice Augustin,et al.  Measuring Multipath Routing in the Internet , 2011, IEEE/ACM Transactions on Networking.

[16]  Yuval Shavitt,et al.  Quantifying the Importance of Vantage Points Distribution in Internet Topology Measurements , 2009, IEEE INFOCOM 2009.