1. Slide -1- February, 2006 Interdomain Routing Gordon Wilfong Distinguished Member of Technical Staff Algorithms Research Department Mathematical and Algorithmic Sciences Center Bell Laboratories February, 2006

2. Slide -2- February, 2006 Current Internet Routing Autonomous Systems (ASes) –Subnetwork under single administrative control Each AS has its own economic incentives to: –cooperate so as to achieve connectivity and to abide by economic contracts (SLAs) –to hide information about its internal network Routing between ASes achieved by protocol where –AS selectively announces to neighbors its chosen route to destinations –AS selects highest ranked route announced to it

3. Slide -3- February, 2006 Border Gateway Protocol (BGP) “I route directly to d” AS1AS2 “The route I use to d goes through AS0” AS0 d “The route I use to d goes through AS1 then AS0” AS3 Data packets

4. Slide -4- February, 2006 Routing Protocol Requirements Protocol must scale –Routing table entries (destinations): ~150,000 and increasing –Autonomous Systems: ~20,000 and increasing Administrators require expressiveness for routing policies –Local AS policy (based on SLAs, traffic engineering etc) determines route selection (effects where AS sends packets) route announcements (effects how/if packets from other ASes traverse it)

5. Slide -5- February, 2006 External BGP (E-BGP) is the mode of BGP that propagates routes between autonomous systems. E-BGP

6. Slide -6- February, 2006 Stable Paths Problem 1 3 2 3 2 0 3 1 0 3 2 1 0 1 0 1 2 0 0 2 1 0 2 0 1 0 210 310 0

7. Slide -7- February, 2006 1 2 0 1 0 Independent policies can lead to nonconvergence 0 1 2 3 2 3 0 2 0 3 1 0 3 0 Bad Gadget Each AS chooses “favorite” available route. 0

8. Slide -8- February, 2006 May Be Multiple Solutions First solution 1 0 2 1 2 0 1 0 1 0 2 1 0 2 2 1 0 2 0 1 2 0 1 0 2 1 0 2 0 1 2 0 1 0 2 1 0 2 0 Second solution DISAGREE

9. Slide -9- February, 2006 What can be done? Modify protocol –Design a protocol that has provably good properties. e.g., [Griffin, W.] Constrain topology and/or policy –Prove good protocol behavior assuming constrained configurations. e.g., [Gao, Rexford] These constraints give rise to interesting “valley-free” routing problems (from customer to local provider to regional provider to national provider and then down chain again to another customer). e.g. [Erlebach et al]

10. Slide -10- February, 2006 Possible Research Directions Complexity issues –What is the complexity of deciding if a configuration is “safe” (i.e., will it always converge)? NP-complete to determine if there is a solution it can converge to [Griffin, W.] Extending model –Model assumes destinations can be considered independently Aggregation has not been taken into consideration Trade-offs between protocol expressivity and safety –Formal model to study such trade-offs

11. Slide -11- February, 2006 Internal BGP (I-BGP) is the protocol used to propagate external routes within an autonomous system. I-BGP

12. Slide -12- February, 2006 Fully Meshed RR A router only announces its own routes

13. Slide -13- February, 2006 Route Reflectors RR Route Reflectors must be fully meshed Route Reflectors pass along updates to client routers

14. Slide -14- February, 2006 Route Selection Summary Highest Local Preference Shortest ASPATH Lowest MED (if same next AS) I-BGP < E-BGP Lowest IGP cost to BGP egress Lowest router ID Traffic engineering Enforce relationships Throw up hands and break ties

15. Slide -15- February, 2006 What Are MEDs? MED=2 MED=0 AS1 AS2 CP Client “Hot Potato Routing” vs. “Cold Potato Routing” MED=1 AS3

16. Slide -16- February, 2006 Persistent Route Oscillations 135.104.54.4 50 40 120 1 MED=1 MED=2 MED=1 r1 r2 r3 A B

17. Slide -17- February, 2006 Oscillations in I-BGP 135.104.54.4 1 1 1 5 5 5 RR1 RR2 RR3 C1 C2C3 P1 P2P3 I-BGP connection Physical link AS0

18. Slide -18- February, 2006 Deflections 135.104.54.4 3 1 1 P Q RR AB Packets can be diverted out of a network unexpectedly. AS0

19. Slide -19- February, 2006 Routing Loops 135.104.54.4 5 1 1 1 RR1 RR2 C1 C2 Badly configured networks can also experience routing loops AS0

20. Slide -20- February, 2006 A Real-world MED Oscillation Example A functioning network breaks into a state of persistent route oscillations when a BGP session goes down First thought to be a hardware problem Analysis shows that route oscillations caused by the use of the MED attribute The example was actually predicted by the theoretical analysis [Griffin, W.]

21. Slide -21- February, 2006 Initial State AS 4 AB EF AS 3 AS 1 AS 2 reflector I-BGP E-BGP (1)(2) C MEDs IGP1 2 D 21 (2) 70000 Only AS 2 sends MEDs to AS 4

22. Slide -22- February, 2006 Initial Routing AS 4 AB EF AS 3 AS 1 AS 2 reflector I-BGP E-BGP (1) C IGP1 2 D 21 (2) MEDs (2) 70000 D prefers AS 2 path due to router ID tie breaking

23. Slide -23- February, 2006 B Changes Its Route AS 4 AB EF AS 3 AS 1 AS 2 reflector I-BGP E-BGP (1) C IGP1 2 D 21 session down! The AS 4  AS 3 BGP Session is dropped (2) MEDs (2) 70000

24. Slide -24- February, 2006 A Changes Its Route AS 4 AB EF AS 3 AS 1 AS 2 reflector I-BGP E-BGP (1) C IGP1 2 D 21 session down! (2) MEDs (2) The MED 1 route from B beats the MED 2 routes that A sees from its clients…. 70000 CALL THIS STATE ZERO

25. Slide -25- February, 2006 C & D Change Routes AS 4 AB EF AS 3 AS 1 AS 2 reflector I-BGP E-BGP (1) C IGP1 2 D 21 session down! (2) MEDs (2) The MED 1 route from A knocks both MED 2 routes out of the picture for C & D … 70000

26. Slide -26- February, 2006 A Changes Route Again AS 4 AB EF AS 3 AS 1 AS 2 reflector I-BGP E-BGP (1) C IGP1 2 D 21 session down! (2) MEDs (2) A now sees the route from D through AS 3, and it is closer IGP-wise than the route from B… 70000

27. Slide -27- February, 2006 C&D Return to Initial Routes AS 4 AB EF AS 3 AS 1 AS 2 reflector I-BGP E-BGP (1) C IGP1 2 D 21 session down! (2) MEDs (2) C & D no longer see MED 1 route from A, so they return to the eBGP routes with MED 2… 70000

28. Slide -28- February, 2006 Back to State Zero! AS 4 AB EF AS 3 AS 1 AS 2 reflector I-BGP E-BGP (1) C IGP1 2 D 21 session down! (2) MEDs (2) A switches back to MED 1 route through B. 70000

29. Slide -29- February, 2006 What Can Be Done? Possible Approaches:  Use only configurations that guarantee no problems  No modification to BGP required  Previous example shows this might be difficult  Prevent problems for any configuration  Modification to BGP required, e.g. [Basu et al]  Minimize overhead (number of messages, memory requirements, …)

30. Slide -30- February, 2006 BGP is extremely flexible, allowing operators to easily make obscure errors that are difficult to find and correct. Policy based routing is difficult to get right. Defining a formal model of path vector protocols helps to understand what features can cause problems and what can be done to avoid errors. Conclusions

31. Slide -31- February, 2006 BGP is the routing protocol in today’s internet What about the future? new Internet architectures new architectures within an AS (e.g., Routing Control Platform [J. Rexford, et al]) In any case, a formal theoretical approach is essential to getting it right. What about the future?

32. Slide -32- February, 2006 Extras

33. Slide -33- February, 2006 Persistent Route Oscillations 135.104.54.4 50 40 120 1 MED=1 MED=2 MED=1 r1 r2 r3 A B r2 r1 r3 r1 r3 r1 r2 r3 r3 r2 r1 r2 r3 r1 r3 r2 r1 3 r1 r3 r2 r1r r3 r3 r2