sibyl a practical internet route oracle
play

Sibyl: A Practical Internet Route Oracle Ethan Katz-Bassett - PowerPoint PPT Presentation

1 Sibyl: A Practical Internet Route Oracle Ethan Katz-Bassett (University of Southern California) with: Pietro Marchetta (University of Napoli Federico II), Matt Calder, Yi-Ching Chiu (USC), Italo Cunha (UFMG), Harsha Madhyastha


  1. 1 Sibyl: A Practical Internet Route Oracle Ethan Katz-Bassett (University of Southern California) with: Pietro Marchetta (University of Napoli Federico II), 
 Matt Calder, Yi-Ching Chiu (USC), Italo Cunha (UFMG), 
 Harsha Madhyastha (Michigan), Vasileios Giotsas (CAIDA) Supported By:

  2. 2 Sibyl: A Practical Internet Route Oracle Ethan Katz-Bassett (University of Southern California) with: Pietro Marchetta (University of Napoli Federico II), 
 Matt Calder, Yi-Ching Chiu (USC), Italo Cunha (UFMG), 
 Harsha Madhyastha (Michigan), Vasileios Giotsas (CAIDA) Supported By:

  3. Traceroute Widely Used 
 by Operators and Researchers 3 “The number one go-to tool is traceroute. � NANOG Network operators troubleshooting tutorial, 2009. ! Lots of use cases: ! topology and AS relationships ! route performance and inflation ! location of congestion ! outages ! prefix hijacks ! etc

  4. Traceroute Widely Used 
 by Operators and Researchers 4 “The number one go-to tool is traceroute. � NANOG Network operators troubleshooting tutorial, 2009. ! Lots of use cases: ! Lots of vantage points: ! topology and AS relationships ! PlanetLab, Ark ! route performance and inflation ! RIPE Atlas, BISmark ! location of congestion ! traceroute servers ! outages ! MobiPerf, Dasu ! prefix hijacks ! RIPE RIS, RouteViews ! etc ! etc

  5. Traceroute Is Extremely Limited 5 “The number one go-to tool is traceroute. � NANOG Network operators troubleshooting tutorial, 2009. ! Lots of use cases: ! Lots of vantage points: ! topology and AS relationships ! PlanetLab, Ark ! route performance and inflation ! RIPE Atlas, BISmark ! location of congestion ! traceroute servers ! outages ! MobiPerf, Dasu ! prefix hijacks ! RIPE RIS, RouteViews ! etc ! etc ! But traceroute only supports one query: 
 “What is the path from vantage point s to destination d?”

  6. 6

  7. How I’d Like to Use Vantage Points 7 Query for: ! Path from a certain network What’s the path from 
 AT&T mobile in LA 
 to YouTube? Current path Historical path

  8. How I’d Like to Use Vantage Points 8 Query for: ! Path from a certain network ! Historical path Did paths from 
 AT&T mobile in LA 
 always go to Seattle? MobiPerf Current path Historical path

  9. How I’d Like to Use Vantage Points 9 Query for: ! Path from a certain network ! Historical path ! Paths through series of hops Do any paths through 
 AT&T in LA to YouTube 
 still go to LA server? MobiPerf Current path Historical path

  10. How I’d Like to Use Vantage Points 10 Query for: ! Path from a certain network ! Historical path ! Paths through series of hops ! Relationship between historical path and current path Other paths that went 
 to YouTube LA and now 
 go to YouTube Seattle? Current path Historical path

  11. How I’d Like to Use Vantage Points 11 Query for: ! Path from a certain network ! Historical path MeasrDroid ! Paths through series of hops ! Relationship between historical path and current path Current path Historical path

  12. How I’d Like to Use Vantage Points 12 Query for: ! Path from a certain network ! Historical path ! Paths through series of hops MeasrDroid ! Relationship between historical path and current path YouTube appears to 
 map these clients 
 together. 
 MobiPerf Current path Historical path

  13. How I’d Like to Use Vantage Points 13 Query for: ! Path from a certain network ! Historical path ! Paths through series of hops MeasrDroid ! Relationship between historical path and current path Or…other paths 
 traversing AT&T -NTT? 
 MobiPerf Current path Historical path

  14. How I’d Like to Use Vantage Points 14 Query for: ! Path from a certain network ! Historical path ! Paths through series of hops MeasrDroid ! Relationship between historical path and current path Or…other paths 
 traversing GTT -NTT? 
 MobiPerf Current path Historical path

  15. How I’d Like to Use Vantage Points 15 Query for: ! Path from a certain network ! Historical path ! Paths through series of hops MeasrDroid ! Relationship between historical path and current path Or…other paths 
 traversing NTT 
 but not AT&T or GTT? 
 MobiPerf Current path Historical path

  16. How I’d Like to Use Vantage Points 16 Query for: ! Path from a certain network ! Historical path ! Paths through series of hops MeasrDroid ! Relationship between historical path and current path Or…other paths 
 traversing NTT 
 but not LA or SEA? 
 MobiPerf Current path Historical path

  17. 2014 Measurement vs 2016 Measurement 17 ! What I do ! What I want to do Experiment Experiment Give me paths like X. Here are some paths. Dest Path Unified Probing Platform Dest1 Path Dest2 MeasrDroid RIPE Atlas 1 RIPE Atlas 2 MobiPerf 1 MobiPerf 2 RIPE Atlas 1 RIPE Atlas 2 MobiPerf 1 MobiPerf 2 MobiPerf 2 MobiPerf 2

  18. Benefit of Combining Platforms 18 100 % of ASes Hosting Vantage Points RIPE+PL +TS+Dasu RIPE 80 TS Dasu 60 PL 40 20 0 10 -1 10 0 10 1 10 2 10 3 10 4 Minimum Customer Cone Size ! Combining platforms improves coverage

  19. Challenge of Combining Platforms 19 1 PL 0.8 CDF of Destinations 0.6 0.4 0.2 0 0 100 200 300 400 500 600 700 800 Number of ASes seen on Path ! Combining platforms improves coverage

  20. Challenge of Combining Platforms 20 1 RIPE 0.8 CDF of Destinations PL 0.6 0.4 0.2 0 0 100 200 300 400 500 600 700 800 Number of ASes seen on Path ! Combining platforms improves coverage

  21. Challenge of Combining Platforms 21 1 RIPE 0.8 CDF of Destinations PL Rate Limited - RIPE 0.6 0.4 0.2 0 0 100 200 300 400 500 600 700 800 Number of ASes seen on Path ! Combining platforms improves coverage ! … but exhaustive probing is infeasible

  22. Challenge of Combining Platforms 22 1 RIPE 0.8 Rate Limited CDF of Destinations - TS+RIPE+PL PL 0.6 Rate Limited - RIPE 0.4 0.2 0 0 100 200 300 400 500 600 700 800 Number of ASes seen on Path ! Combining platforms improves coverage ! … but exhaustive probing is infeasible

  23. Challenge of Combining Platforms 23 1 RIPE 0.8 CDF of Destinations Rate Limited(G) - TS+RIPE+PL Rate Limited 0.6 - TS+RIPE+PL PL Rate Limited(G) 0.4 - RIPE Rate Limited - RIPE 0.2 0 0 100 200 300 400 500 600 700 800 Number of ASes seen on Path ! Combining platforms improves coverage ! … but exhaustive probing is infeasible ! Rate limits mean you have to be smart about what to issue

  24. Goals 24 ! Take advantage of diverse vantage points ! Efficient use of probing budgets ! High rate, low diversity vantage points like Ark and PL ! Low rate, high diversity vantage points like Atlas and LG ! Support rich queries Service queries: ! as if we had traceroutes from all vantage points to all Internet destinations ! even though probing budgets are very restricted

  25. Related Work 25 ! IXPs: Mapped?, RocketFuel, Lord of the Links, Reverse Traceroute, etc ! Issued measurements likely to traverse particular links ! iPlane ! Predicted source/destination paths ! TopHat ! Unified historical and current ! Multiple testbeds? ! Does it exist anymore? ! MPlane ! Srikanth’s talk ! Others?

  26. Goals 26 ! Take advantage of diverse vantage points ! Efficient use of probing budgets ! High rate, low diversity vantage points like Ark and PL ! Low rate, high diversity vantage points like Atlas and LG ! Support rich queries Service queries: ! as if we had traceroutes from all vantage points to all Internet destinations ! even though probing budgets are very restricted

  27. Optimize Use of Probing Budget 27 Goal: ! In each round, allocate probing budget to best serve queries

  28. Optimize Use of Probing Budget 28 Goal: ! In each round, allocate probing budget to best serve queries

  29. Optimize Use of Probing Budget 29 Goal: ! In each round, allocate probing budget to best serve queries ! Max utility of traceroutes

  30. Optimize Use of Probing Budget 30 Goal: ! In each round, allocate probing budget to best serve queries ! Max utility of traceroutes ! T r : set of traceroutes in the round is union of those from each platform V

  31. Optimize Use of Probing Budget 31 Goal: ! In each round, allocate probing budget to best serve queries ! Max utility of traceroutes ! T r : set of traceroutes in the round is union of those from each platform V ! Utility of set of traceroutes T r is sum of utility in matching each query q

  32. Optimize Use of Probing Budget 32 Goal: ! In each round, allocate probing budget to best serve queries ! Max utility of traceroutes ! T r : set of traceroutes in the round is union of those from each platform V ! Utility of set of traceroutes T r is sum of utility in matching each query q ! Each platform V has a rate limit budget C V

  33. Challenges in Optimization 33 Goal: ! In each round, allocate probing budget to best serve queries Need to solve (rest of talk): ‣ What is query language? 
 ‣ How to find paths to match queries, given we can’t issue every traceroute (or know the results before issuing)? 
 ‣ How to solve the optimization efficiently?

Recommend


More recommend