1 / 27

Evaluating Potential Routing Diversity for Internet Failure Recovery

Evaluating Potential Routing Diversity for Internet Failure Recovery. *Chengchen Hu, + Kai Chen, + Yan Chen, *Bin Liu *Tsinghua University, + Northwestern University. Internet Failures. Failure is part of everyday life in IP networks

milek
Télécharger la présentation

Evaluating Potential Routing Diversity for Internet Failure Recovery

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Evaluating Potential Routing Diversity for Internet Failure Recovery *Chengchen Hu, +Kai Chen, +Yan Chen, *Bin Liu *Tsinghua University, +Northwestern University

  2. Internet Failures Failure is part of everyday life in IP networks e.g., 675,000 excavation accidents in 2004 [Common Ground Alliance] Network cable cuts every few days … Real-world emergencies or disasters can lead to substantial Internet disruption Earthquakes Storms Terrorist incident: 9.11 event …

  3. Example: Taiwan earthquake incident Large earthquakes hit south of Taiwan on 26 December 2006 Only two of nine cross-sea cables not affected There were still abundant physical level connectivity there, but it took too long for ISPs to find them and use them. 3 figures cited from "Aftershocks from the Taiwan Earthquakes: Shaking up Internet transit in Asia, NANOG42"

  4. How reliable the Internet is? • Internet is not as reliable as people expected! [Wu, CoNEXT’07] • 32% ASes are vulnerable to a single critical customer-provider link cut • 93.7% Tier-1 ISP’s single-homed customers are lost from the peered ISP due to Tier-1 depeering • Our question: can we find more resources to increase the Internet reliability especially when Internet emergency happens?

  5. Roadmap • Background • Where are the potential resources? • How much potential resources are there? • How to use the potential resources?

  6. Basic Idea • Two places where we can find more routing diversities: • Internet eXchange Points (IXPs) • Co-location where multiple ASes exchange their traffic • Participant ASes in an IXP may not be connected via BGP • Internet valley-free routing policy • AS relationships: customer-provider, peering, sibling • Peering relaxation (PR): allow one AS to carry traffic from the other to its provider • Mentioned in [Wu, CoNEXT’07], but no evaluation • Our main focus: • How much can we gain from these two potential resources, i.e., IXP and PR?

  7. Roadmap • Background • Where are the potential resources? • How much help could provide? • How to use the potential resources?

  8. Dataset for Evaluation • Most complete AS topology graph • BGP data • Route Views, RIPE/RIS, Abilene, CERNET BGP View • P2P traceroute • Traceroute data from 992, 000 IPs in over 3, 700 ASes • In total, 120K AS links with AS relationships • http://aqualab.cs.northwestern.edu/projects/SidewalkEnds.html [Chen et al, CoNEXT’09] • IXP data • PCH + Peeringdb + Euro-IX (~200 IXPs) • 3468 participant ASes

  9. Failure Models • Tier-1 depeering • Real example: Cogent and Level3 depeering • Tier-1 provider-customer link teardown • Reported in NANOG forum • Mixed types of link breakdown • 9.11 event, Taiwan earthquakes, 2003 Northeast blackout

  10. Evaluation Metrics • Recovery Ratio • # of recovered <src-dst> AS pairs versus total # of affected <src-dst> AS pairs • Path Diversity • # of increased link-disjoint AS paths between affected <src-dst> AS pairs • Shifted Path • # of link-disjoint AS paths shifted onto a normal link after we use IXP or PR resources More results please check our report "Evaluating Potential Routing Diversity for Internet Failure Recovery" @ http://s-router.cs.tsinghua.edu.cn/pub/IER_report.pdf

  11. Results: Tier-1 Depeering • 36 experiments for 9 Tier-1 ASes • Recovery ratio: most of the lost AS pairs can be recovered

  12. Results: Tier-1 Depeering • Path diversity: multiple AS paths between lost AS pairs

  13. Results: Tier-1 Depeering • Shifted path • On average, 3.75 ~ 17.2 for all 36 experiments • Moderate traffic load shifted onto the unaffected links

  14. Roadmap • Background • Where are the potential resources? • How much help could provide? • How to use the potential resources?

  15. Economic model • B pays to A for recovery • Business model • Risk alliance (like airlines): price is determined beforehand • pay on bandwidth & duration or bits (95 percentile) peer A B IXP A B P-C A B P-C A B

  16. Communication channel • Search for peers • Have direct connections to peers • Search for co-located ASes in the same IXP • ASes are connected by switches in modern IXPs • Messages are broadcasted via switches • Message confidentiality through public key crypto

  17. Automatic communications: query phase potential helper A who can connect to specific destination ASes? potential helper B query victim AS potential helper C potential helper D

  18. Automatic communications: Check availability 1. Check connectivity (traceroute) 2. Check available bandwidth (IGI/Yaz/pathload/spruce) potential helper A potential helper B victim AS potential helper C potential helper D

  19. Automatic communications: reply phase • Message confidentiality with public key crypto potential helper A • I can provide X bandwidth to it potential helper B reply victim AS potential helper C reply • I can provide Y bandwidth to it potential helper D

  20. Automatic communications: ACK phase potential helper A • I would like buy Z (<=Y) potential helper B victim AS ACK potential helper C potential helper D

  21. Automatic communications: new BGP session potential helper A potential helper B victim AS Set up BGP session Can be withdrawn later potential helper C potential helper D

  22. Optimal selection of helper ISPs • From a single victim ISP perspective • Buy transit from a minimal number of ASes • Recover all the (prioritized) traffic • Least cost

  23. Selection heuristic is how much bandwidth AS j could provide to Di; Lost connectivity to {Di}, with bandwidth demand {Bi}

  24. Selection heuristic Score each (helper) AS j with Select the AS with largest score (select the one with lowest price if same score) Lost connectivity to {Di}, with bandwidth demand {Bi} 3 2.3 5 2.1

  25. Selection heuristic updated Update Lost connectivity to {Di}, with bandwidth demand {Bi}

  26. Selection heuristic rescore and select Lost connectivity to {Di}, with bandwidth demand {Bi} 0.3 1 0 0.1

  27. Summary Point out a new venue for Internet failure recovery. Evaluate the potential routing diversity via IXP and PR with the most complete AS topology graph. 40%-80% of affected <Src, Dst>AS pairs can be recovered via IXP and PR with multiple paths and moderate shifted paths. Possible and practical mechanisms to utilize potential routing diversity. Look forward to feedback and collaborations from IXP/ISPs!

More Related