1. 1 Network Topology Measurement Yang Chen CS 8803

2. 2 Outline Big Picture ISP Topology Measurement Statistical Results Problems & Solutions

3. 3 Heuristics for Internet Map Discovery R. Govindan and H. Tangmunarunkit INFOCOM 2000

4. 4 Why do we need the topology? Understand the macroscopic properties of the Internet physical structure Network management Topology-aware algorithms Simulation and topology generation tools

5. 5 On-going efforts CAIDA SkitterRouter View

6. 6 Fundamental: traceroute Prober sends packets with successively increased TTL. A router responds with ICMP time exceeded when the probe is with TTL=1

7. 7 Fundamental: traceroute Geographic info can help on building up the topology. * Data from http://www.linkwan.com/vr/

8. 8 Fundamental: Tree => map 1)Source routing 2)Multiple Vantage points

9. 9 Address scan space BGP tables Route Table Database Informed Random Address Probing –A response from some IP address is considered as a sign that some prefix P of A must contain addressable nodes; –If P is an addressable prefix, the neighboring prefixes of P are also considered as possibly addressable. (128.8/16 and 128.10/16 are neighbors of 128.9/16)

10. 10 Some results 150,000 interfaces and nearly 200,000 links Findings related to source route –Simulation demonstrated that In relatively sparse random networks, a few source route capable nodes (< 5%) are sufficient to discover 90% of the links. In fact, there are 8% routers support source route. –Source route discovered links do not skew the qualitative conclusion on the network statistics.

11. 11 For example: degree distribution Similar observation on hop-pair distribution

12. 12 Measuring ISP Topologies with Rocketfuel N.Spring, R. Mahajan and D. Wetherall ACM Sigcomm 2002

13. 13 ISP network infrastructure Access Router Backbone Router Backbone Link

14. 14 ISP topology measurement An old story: the blinds and the elephant ISP Traceroute

15. 15 ISP topology measurement Traceroute Server

16. 16 Focusing on one ISP – Directed probing Network Next Hop M/LP/Weight Path * 192.9.9.0 204.212.44.128 0 234 266 237 3561 701 90 i * 205.238.48.3 0 2914 1 90 i * 144.228.240.93 0 1239 701 90 i * 204.70.4.89 0 3561 1 90 i * 194.68.130.254 0 5459 5413 1 90 i *> 134.24.127.3 0 1740 701 90 i * 202.232.1.8 0 2497 701 90 i * 158.43.133.48 0 1849 702 701 90 i * 131.103.20.49 0 1225 2548 1 90 i blackrose.org (Ann Arbor) 204.212.44.128 through AS234 Verio (MAE-WEST) 205.238.48.3 through AS2914 Sprint (Stockton) 144.228.240.93 through AS1239 MCI (San Francisco) 204.70.4.89 through AS3561 LINX (London) 194.68.130.254 through AS5459 CERFnet (San Diego) 134.24.127.3 through AS1740 IIJ (Japan) 202.232.1.8 through AS2497 PIPEX (London) 158.43.133.48 through AS1849 IAGnet (Chicago) 131.103.20.49 through AS1225 * BGP table source: RouteView project

17. 17 Focusing on one ISP – Directed probing Traceroutes to dependent prefixes: All traceroutes to these prefixes from any vantage point should transit the ISP. Dependent prefixes can be readily identified from the BGP table. All AS-paths for the prefix would contain the number of the AS being mapped. Traceroutes from insiders: We call a traceroute server located in a dependent prefix an insider. Traceroutes from insiders to any prefix should transit the ISP. Traceroutes that are likely to transit the ISP based on some AS-path are called up/down traces.

18. 18 Path/Query reduction Share IngressShare egress Same next-hop AS number

19. 19 Impacts of directed probing 1)Fraction of useful but pruned traces from 0.1 to 7% 2)Unnecessary traces around 6% over all the ISPs * Comparison based on Skitter data

20. 20 Impacts of ingress reduction Overall, ingress reduction keeps only 12% of the traces chosen by directed probing. The number of vantage points that share an ingress by rank

21. 21 Impacts of egress reduction The number of dependent prefixes that share an egress by rank Overall, egress reduction keeps only 18% of the Dependent prefix traces chosen by directed probing.

22. 22 Impacts of next-hop reduction Overall, Next-hop AS reduction Reduces the number of traces to 5% of those chosen by directed probing.

23. 23 POP sizes analysis

24. 24 Power Law Complementary cumulative distribution function (CCDF) P(X>x) Pareto Distribution Power Law

25. 25 Router degree distribution

26. 26 Peering structure

27. 27 Difficulties in topology discovery Shared media Backup links Router Identification and annotation Alias resolution Completeness Validation Currently, none of them is completely solved!

28. 28 POP hierarchy Naming convention, DNS information and neighbor inferring

29. 29 Backbone topology AT&T Level 3

30. 30 Alias Problem OR

31. 31 Alias: is it a big deal?

32. 32 Alias resolution Send a packet with unreachable port to certain interfaces which are possible alias. The corresponding ICMP port unreachable response will contain the source address. IP identifier

33. 33 Completeness validation Comparison with Router Views Comparison with Skitter IP address space –Search prefixes of ISP’s address space for additional IP addresses Validation with ISPs –Is “Good” enough?

34. 34 In Search of Path Diversity in ISP Networks P. Teixeira, K. Marzullo, S. Savage and G. M. Voelker IMC 2003

35. 35 Real metric instead of counting links Path diversity –Metric that reflects the number of routes available between two points in the network An extreme example

36. 36 Real topology speaks Inter-PoP Path diversity in the Sprint Network Inter-PoP Path diversity inferred by Rocketfuel

37. 37 Take a closer look

38. 38 Inaccuracy introduced during probing Lack of vantage points –How many points are sufficient? Incomplete traceroutes –What can we do if ISP turns off traceroute functionality? Changes in the path of a probe Incorrect DNS record

39. 39 Inaccuracy from processing probed links Alias Resolution Adding reverse links Missed and added links in Rocketfuel PoP topology relative to the number of links in the Sprint real topology

40. 40 Questions?