BGP Beacons, Network Tomography, and Bayesian Computation to Locate Route Flap Damping

Gray, Caitlin; Mosig, Clemens; Bush, Randy; Pelsser, Cristel; Roughan, Matthew; Schmidt, Thomas C.; Wählisch, Matthias

doi:10.1145/3419394.3423624

Cited by 16 publications

(11 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since these use cases either rely on route engineering or on understanding route changes, revtr 2.0 should be able to measure these routes in roughly the time period that traffic engineering techniques are given to converge. Announcements every 15 minutes were subject to route flap damping [41], meaning that faster changes impede traffic engineering, and so we want to support quickly bootstrapping outside sources and issuing ≈15,000 reverse traceroutes within 15 minutes. This rate equates to ≈1.4M per day, but, in practice, most studies are more conservative (waiting 90 [82] or 120 [19,25,53,66] minutes between announcements), and so 1.4M per day will suffice to support multiple parallel uses.…”

Section: Goalsmentioning

confidence: 99%

“…revtr 2.0 supports quick iteration: Reverse traceroute measurements took 9-13 minutes per routing configuration, similar to the other required steps: 15 minutes to wait for BGP convergence and avoid route flap dampening [36,41,53,71,82] and 15 minutes to refresh the atlas. revtr 2.0 can be integrated into automated TE systems, providing scale and coverage that cannot be matched by querying operators or looking glasses.…”

Section: Supporting Traffic Engineeringmentioning

confidence: 99%

See 1 more Smart Citation

Internet scale reverse traceroute

Vermeulen¹,

Gurmericliler

Cunha

et al. 2022

Proceedings of the 22nd ACM Internet Measurement Conference

View full text Add to dashboard Cite

Knowledge of Internet paths allows operators and researchers to better understand the Internet and troubleshoot problems. Paths are often asymmetric, so measuring just the forward path only gives partial visibility. Despite the existence of Reverse Traceroute, a technique that captures reverse paths (the sequence of routers traversed by traffic from an arbitrary, uncontrolled destination to a given source), this technique did not fulfill the needs of operators and the research community, as it had limited coverage, low throughput, and inconsistent accuracy. In this paper we design, implement and evaluate revtr 2.0, an Internet-scale Reverse Traceroute system that combines novel measurement approaches and studies with a large-scale deployment to improve throughput, accuracy, and coverage, enabling the first exploration of reverse paths at Internet scale. revtr 2.0 can run 15M reverse traceroutes in one day. This scale allows us to open the system to external sources and users, and supports tasks such as traffic engineering and troubleshooting. CCS CONCEPTS• Networks → Network measurement; Network monitoring.

show abstract

Section: Goalsmentioning

confidence: 99%

Section: Supporting Traffic Engineeringmentioning

confidence: 99%

Internet scale reverse traceroute

Vermeulen¹,

Gurmericliler

Cunha

et al. 2022

Proceedings of the 22nd ACM Internet Measurement Conference

View full text Add to dashboard Cite

show abstract

“…Path-to-link tomography can also enable efficient estimation of link metrics even when there are no such restrictions [26], [27]. BeCAUSe [28] is an interesting recent work that uses path-to-link tomography to reveal the deployment of route flap damping in BGP routers. Network troubleshooting (identifying and locating faults) is yet another application.…”

Section: B Related Workmentioning

confidence: 99%

PAINT: Path Aware Iterative Network Tomography for Link Metric Inference

Xue

Marina

et al. 2022

2022 IEEE 30th International Conference on Network Protocols (ICNP)

View full text Add to dashboard Cite

Understanding link-level performance is key to assuring the quality of cloud-based and OTT services, optimal path selection, robust network operations and beyond. However, direct measurement of each link not only incurs high overhead at the Internet-scale but also is infeasible due to lack of access to network measurement information beyond AS boundaries and functional limitations at relay nodes. Although network tomography is well suited, existing approaches are insufficient due to their unrealistic assumptions with respect to stability, controllability, and visibility. Motivated by this, we propose PAINT, an online iterative algorithm that estimates and refines linklevel performance metrics based on path-level measurement. In PAINT, the link metrics are iteratively estimated by minimizing their least square error (LSE) and calibrated based on the comparison of weight between the estimated shortest paths (SPs) and best-known paths from end-to-end path measurements. The key insight is that when there is inconsistency between these paths, then weights of links on the estimated SP are likely misestimated, triggering a further round of estimation to refine the estimated link metrics. Evaluation of PAINT, focusing on link delay estimation, using four different real network topologies and two real-world measurement datasets (including one we collected) shows that relative to existing approaches, it yields up to 3x gain in absolute link delay estimation accuracy and improves decisions dependent on link delay estimation by up to 5x in relative error.

show abstract

“…These variations are mainly due to path hunting and noise reduction techniques (e.g. MRAI and Route Flap Damping [4,6,8,12,19]) and are hardly predictable. (3) Local topological changes.…”

Section: Hunting Bgp Zombiesmentioning

confidence: 99%

“…Understanding prefix withdrawal propagation time is key to differentiate both events. Past research has shown that withdrawals usually lasts a few minutes [11] and in the case of Route Flap Damping up to one hour [8]. From our data set we estimate typical withdrawing time, , by looking at the time duration of ( ) drops, that is − , necessary for prefix to be completely withdrawn ( ( ) = 0 and ( ) > ℎ ).…”

Section: Withdrawal Propagation Timementioning

confidence: 99%

Hunting BGP zombies in the wild

Ongkanchana

Fontugne

Esaki

et al. 2021

Proceedings of the Applied Networking Research Workshop

View full text Add to dashboard Cite

As the key component of Internet's inter-domain routing, BGP is expected to work flawlessly. However, a recent study has revealed the presence of BGP zombies: Withdrawn prefixes that are still active in routing tables and that can cause routing issues. That study used experimental prefixes with scheduled withdrawals (BGP beacons). In this study we aim at detecting BGP zombies for any prefixes announced on the Internet. To that end we study characteristics of withdrawn messages, and devise a method to differentiate withdraw messages corresponding to local topological changes to those standing for prefixes withdrawn by their origin AS. Based on this classification we study the occurrence of zombies in the wild in six years of BGP data. We find over 6.5 millions zombies, among those we confirm that 94% report incoherent states and caused 468 potential routing loops. Our study also reveals that noisy prefixes, long AS paths, and ASes announcing a large number of prefixes are more prone to zombies.

show abstract

BGP Beacons, Network Tomography, and Bayesian Computation to Locate Route Flap Damping

Cited by 16 publications

References 27 publications

Internet scale reverse traceroute

Internet scale reverse traceroute

PAINT: Path Aware Iterative Network Tomography for Link Metric Inference

Hunting BGP zombies in the wild

Contact Info

Product

Resources

About