1 / 40

Optimising Path Discovery Doubletree and Paris Traceroute

Optimising Path Discovery Doubletree and Paris Traceroute. Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz. About Finding The Path between Two Monitors. Traceroute. Traceroute (the real story) ‏.

Download Presentation

Optimising Path Discovery Doubletree and Paris Traceroute

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. Optimising Path DiscoveryDoubletree and Paris Traceroute • Tony McGregor • RIPE NCC Visiting Researcher • tony.mcgregor@ripe.net • The University of Waikato • tonym@cs.waikato.ac.nz

  2. About Finding The Path between Two Monitors • Traceroute

  3. Traceroute(the real story)‏

  4. Traceroute(the really real story)‏

  5. Traceroute(more reality)‏

  6. Traceroute(But wait ... there's more)‏ • RFC1918 addresses • ICMP disabled • ICMP filters • False routes • Load balancing • Interface coalition • ISPs hide topology • TTL 0 • Tunnels • MPLS • IPv6

  7. Wasted Probes(doubletree)‏ Donnet B et al. “Efficient Algorithms for Large-Scale Topology Discovery” Sigmetrics 05

  8. Wasted Probes(doubletree)‏

  9. Wasted Probes(doubletree)‏

  10. Wasted Probes(doubletree)‏

  11. Wasted Probes(doubletree)‏

  12. Wasted Probes(doubletree)‏

  13. Wasted Probes(doubletree)‏

  14. Wasted Probes(doubletree)‏

  15. Wasted Probes(doubletree)‏

  16. Wasted Probes(doubletree)‏

  17. Intra-monitor Redundancy

  18. Intra-monitor Redundancy

  19. Intra-monitor Redundancy

  20. Intra-monitor Redundancy

  21. Intra-monitor Redundancy

  22. Extent of intra-monitor redundancy • Donnet et al. Estimate that in skitter 87% of probes were redundant. skitter doubletree

  23. Inter-monitor redundancy

  24. Inter-monitor redundancy

  25. Inter-monitor redundancy

  26. Inter-monitor redundancy

  27. Inter-monitor redundancy skitter doubletree

  28. Limitations • Doubletree finds fewer nodes that repeated traceroutes • Path changes • Load balancing • Global stop set requires communication • Negates savings to some extent • May be large if there are many destinations • Trade off between reducing intra- and inter- monitor redundancy

  29. Doubletree for DAR • The DAR project is investigating the design of a very large scale active measurement system • Design should support in the order of 100.000 probes • Want to find path to a small number of targets from a large number of probes • can't just probe • double tree • A lot of redundancy so much potential for savings • Need to share global stop set between many probes • Global stop set will be smaller than for scamper

  30. Discovering Alternative Pathsload balancing and Paris traceroute • TCP traceroute study (July 2006, 5,000 addresses) • 7 of 9 Tear one ISPs • 17 of 64 top regional ISPs • Over 80% of destinations appeared to have load balancers in their path Augustin et al. “Avoiding traceroute anomalies with Paris traceroute” Université of Pierre et Marie Curie

  31. Alternative Pathsnaïve traceroute

  32. Alternative Pathsnaïve traceroute

  33. Alternative Pathsnaïve traceroute

  34. Alternative Pathsnaïve traceroute

  35. Alternative Pathsnaïve traceroute

  36. Alternative Pathsnaïve traceroute

  37. Alternative Pathsnaïve traceroute

  38. Load Balancing • Three types of load balancers • Random (per-packet)‏ • Destination address based • Flow ID based • Some or all of the classic 5-tuple • TOS • ICMP code and checksum • Traceroute deliberately varies flow information • Match TTL expired with probe packet

  39. Paris Traceroute • Goal: • To discover all paths from a source to a destination • Doesn't address per-destination load balancing • Create headers such that flow ID remains the same • Still need to match ICMP TTL expired packets to probes • For UDP vary checksum • Requires manipulating the payload • For ICMP vary sequence number and identifier • So that checksum remains constant • For TCP vary the sequence number • Uses TCP half-open technique

  40. Results • For a particular set of 5000 destinations from Université Pierre et Marie Curie • Loops • 5.3% of traceroute paths • 87% caused by per flow load balancing • Cycles • 0.84% of traceroute paths • 78% caused by per-flow load balancing • Diamonds • 79% of traceroute paths • 64% cause by per-flow load balancing

More Related