1. Network Protocols: Design and Analysis Polly Huang EE NTU http://cc.ee.ntu.edu.tw/~phuang phuang@cc.ee.ntu.edu.tw

2. Internet Routing III [Tsuchiya88a] [Labovitz00a]

3. Landmark Routing [Tsuchiya88a]

4. Polly Huang, NTU EE4 Context fairly early in the Internet life –before BGP-3 –before CIDR example of SIGCOMM “wild idea” paper

5. Polly Huang, NTU EE5 Key Idea Self-configuring hierarchy for routing with many routers

6. Polly Huang, NTU EE6 Why Landmark Routing? area routing requires knowledge of topology, maybe doesn’t get best aggregration possible LM knows about internal structure of nearby nodes, even if in different AS dynamic address assignment—easier to manage reduce size of routing table… because address are automatic, and reassigned on-demand, can get better aggregation than area hierarchy could be more reliable if congestion because supports multiple (?) different approach than area routing

7. Polly Huang, NTU EE7 Landmark Routing Disadvantages don’t always get shortest path [but true about all routing protocols that have aggregation/policy] admin control? (paper hints at approaches, but not fully explored) performance not fully explored? –less info further away from destination, therefore more likely to get poor quality routes to it [but no different from area routing] –performance of LM placement/config algorithms? combines routing and address (but so does area routing) addressing –address may not be stable –LM uses variable length address

8. Polly Huang, NTU EE8 Landmark hierarchy Details about things nearby and less information about things far away Not defined by arbitrary boundaries –thus, not well suited to the real world that does have administrative boundaries –(although he says something about adding admin boundaries)

9. Polly Huang, NTU EE9 A Landmark 1 2 3 4 5 6 7 8 9 10 11 Router 1 is a landmark of radius 2

10. Polly Huang, NTU EE10 Landmark Overview Landmark routers have “height” which determines how far away they can be seen (visibility) Routers within Radius n can see a landmark router LM(n) See means that those routers have LM(n)’s address and know next hop to reach it. –Router x as an entry for router y if x is within radius of y Distance vector style routing with simple metric Routing table: Landmark (LM2(d)), Level(2), Next hop

11. Polly Huang, NTU EE11 LM Hierarchy Definition Each LM (Li) associated with level (i) and radius (ri) Every node is an L0 landmark Recursion: some Li are also Li+1 –Every Li is seen by at least one Li+1 Terminating state when all level j LMs see entire network

12. Polly Huang, NTU EE12 LM addresses LM(2).LM(1).LM(0) (x.a.b and y.a.b) LM level maps to radius (part of configuration), e.g.: –LM level 0: radius 2 –LM level 1: radius 4 –LM level 2: radius 8 If destination is more than two hops away, will not have complete routing information, refer to LM(1) portion of address, if not then refer to LM(2).. (c would forward based on y in y.a.b) X y a b c

13. Polly Huang, NTU EE13 LM Routing LM does not imply hierarchical forwarding It is not a source route En route to LM(1) may encounter router that is within LM(0) radius of destination address (like longest match) Paths may be asymmetric

14. Polly Huang, NTU EE14 LM self-configuration Bottom-up hierarchy construction algorithm –goal to bound number of children Every router is L0 landmark All routers advertise themselves over a distance All Li landmarks run election to self-promote one or more Li+1 landmarks Dynamic algorithm to adapt to topology changes--Efficient hierarchy

15. Polly Huang, NTU EE15 Landmark Routing: Basic Idea Source wants to reach LM0[a], whose address is c.b.a: Source can see LM2[c], so sends packet towards c Entering LM1[b] area, first router diverts packet to b Entering LM0[a] area, packet delivered to a - Not shortest path - Packet does not necessarily follow specified landmarks

16. Polly Huang, NTU EE16 Landmark Routing: Example

17. Polly Huang, NTU EE17 Routing table for Router g LandmarkLevelNext hop LM2[d] LM0[e] LM1[i] LM0[k] LM0[f] 2 1 0 0 0 f k f k f r0 = 2, r1 = 4, r2 = 8 hops Router g How to go from d.i.g to d.n.t? How does path length compare to shortest path? Router t

18. Polly Huang, NTU EE18 Evaluation analytic results –but bounds not very helpful simulation –routing table size (R) –mean path length –distance to nearby landmark –(seems weak) [Figure 6 from Tsuchiya88a] r/d = radius/distance rtg table size mean path len

19. Questions?

20. BGP Routing Convergence Times [Labovitz00a]

21. Polly Huang, NTU EE21 Context BGP widely deployed in the Internet but poorly understood

22. Polly Huang, NTU EE22 Key Idea convergence time takes longer we expected observes 2-3 minute convergence times (6x longer than expected!) bounds on BGP convergence: O(n!) worst case, O((n-3)*30s) [n is number of ASes]

23. Polly Huang, NTU EE23 Why is Convergence Important? robustness –PSTN (telephone) failover times are in milliseconds –Internet failover times are in 10s of seconds –open research question: how can Internet routing do much better?

24. Polly Huang, NTU EE24 Methodology experiments over Internet: manually injected faults propagate across net simulation to study worst case behavior theoretical analysis—helps understand worst case bounds traces of 2 years of convergence times

25. Polly Huang, NTU EE25 Methodology Picture Internet-scale experimentation. What kinds of complexities arise? Have to be careful with real routes; ([Labovitz00a] Figure 1)

26. Polly Huang, NTU EE26 Short->Long Fail-Over (Tlong) New Route, Long->Short Fail-over (Tup and Tshort) Failure (Tdown) Long tailed distribution (up to 15 minutes); more msgs in longer waits; long absolute times Observed Convergence Latency Labovitz00a Figure 2a

27. Polly Huang, NTU EE27 Other Observations No correlation between network distance (latency, router, or AS hops) and convergence times Why is long convergence bad?…

28. Polly Huang, NTU EE28 Affects on Traffic ([Labovitz00a] figure 4a) Why does loss go up? There’s always a direct path? some people use old paths, routing loops

29. Polly Huang, NTU EE29 How To Tell What’s Going On? Simulate BGP –model one router per AS –assume full routing mesh –ignore latency –synchronous processing via global queue  simple model that captures key details

30. Polly Huang, NTU EE30 What’s going on? there are many possible routes (indirect through other ASes) and it takes a long time w/BGP to figure out that none work –BGP can try all paths of length 2, then 3, then 4 => O(n!) steps –even with min-route-adver it still can take O(n) steps

31. BGP Convergence Example R AS0 AS1 AS2 AS3 *B Rvia 3 B R via 03 B R via 23 *B Rvia 3 B R via 03 B R via 13 *B Rvia 3 B R via 13 B R via 23 AS0AS1AS2 *** *B R via 203 *B R via 013 B R via 103

32. Polly Huang, NTU EE32 What about MinRouteAdver? BGP has a minimum advertisement interval timer –designed to limit updates –and to encourage aggregation How does it affect convergence? –by delaying announcements, routers figure out the pain sooner –see section 5.2 result: n-3 rounds rather than n!

33. Polly Huang, NTU EE33 Does this explain measurements? Tup/Tshort converge quickly because they shorten path length and therefore are quickly accepted Tdown/Tlong converge slowly because BGP tries hard to find all alternatives –Tlong actually sometimes goes quicker if it’s “not long enough” and can preempt some of the thrashing

34. Polly Huang, NTU EE34 Other Observations Could do loop detection at sender side and not just receiver side xxx

35. Questions?