Convergence Xiaoqiang Wang 1 , Olivier Bonaventure 2 , Peidong Zhu 1 - - PowerPoint PPT Presentation

▶
convergence
SMART_READER_LITE
LIVE PREVIEW

Convergence Xiaoqiang Wang 1 , Olivier Bonaventure 2 , Peidong Zhu 1 - - PowerPoint PPT Presentation

Mar 10, 2024 540 likes •811 views

Stabilizing BGP Routing without Harming Convergence Xiaoqiang Wang 1 , Olivier Bonaventure 2 , Peidong Zhu 1 1 National University of Defense Technology, China 2 University Catholique de Louvain, Belgium Outline Background Motivation

slide-1
SLIDE 1

Stabilizing BGP Routing without Harming Convergence

Xiaoqiang Wang1, Olivier Bonaventure2, Peidong Zhu1

1 National University of Defense Technology, China 2 University Catholique de Louvain, Belgium

slide-2
SLIDE 2

Outline

 Background  Motivation  Methodology  Evaluation  Conclusion

2

slide-3
SLIDE 3

Outline

 Background  Motivation  Methodology  Evaluation  Conclusion

3

slide-4
SLIDE 4

Qwest:209 UCLA:52 Sprint:1239 SBC:7132 Autonomous System (AS) B A Internet

Internet Routing

:

Intra-AS: OSPF, IS-IS, RIP Inter-AS: BGP (Border Gateway Protocol)

4

slide-5
SLIDE 5

BGP

 Prefix specific  Path Vector Routing Protocol  One-fits-all-model

AS 1 AS 2 AS 3 AS 4 AS 5 AS 6 AS 7 AS 8 prefix d

5

slide-6
SLIDE 6

BGP Churn

 Large volume of BGP updates  Bad for routers

Overload CPU,memory, frequent FIB changes

 Major causes

BGP path exploration Route flapping

6

slide-7
SLIDE 7

BGP Path Exploration

 Single event triggers several updates

7

AS 1 AS 2 AS 3 AS 4 AS 5 AS 6 AS 7 AS 8 prefix d

slide-8
SLIDE 8

Route Flapping

 Routes periodically change  Reasons are diverse  Mice-elephant

a significant portion of churn is associated to

a small set of highly active prefixes [Rexford 02,

Oliveira05]

3% prefixes  36% updates [Pelsser PAM11]

8

slide-9
SLIDE 9

Current countermeasures

 Path exploration acceleration

BGP-RCN, EPIC Not deployed yet

 Suppress excessive BGP updates

 Route flap Damping, MRAI  Only two built-in mechanisms in real router  Dying for breaking/delaying convergence

9

slide-10
SLIDE 10

Route Flap Damping

 Principle

A penalty per peer and per prefix Update penalty upon receiving an update Suppress a route if associated penalty > T Penalty exponentially decays over time

 Triggered by 3 flaps under cisco

parameter[Mao02, Randy02]

 Interactions between RFD and BGP path

exploration

10

suppressed time

slide-11
SLIDE 11

MRAI

 Minimum Route Advertisement Interval  Supposed to be per peer and per prefix  Rate-limit BGP updates

Two consecutive announcements are spaced

at least a MRAI interval*jitter[0.75,1]

Typical setting: 30s for eBGP, 5s for iBGP

 BGP updates are heavily delayed

11

slide-12
SLIDE 12

Outline

 Background  Motivation  Methodology  Evaluation  Conclusion

12

slide-13
SLIDE 13

Motivation(1/2)

 Analyze BGP change type

Data set: one month updates from RouteView Duplicated updates are filtered per prefix Consider only Announcement messages Compare two adjacent updates

BGP churn mostly caused by AS_PATH and COMMUNITY changes

13

slide-14
SLIDE 14

Motivation(2/2)

 Path Locality

An AS explores limit number of AS_PATHs to

reach highly active prefixes

Same data set as in previous slide for each prefix, we define locality likelyhood

𝑚𝑗𝑙𝑓𝑚𝑧ℎ𝑝𝑝𝑒 =

#{𝑢𝑝𝑞 3 𝑞𝑏𝑢ℎ𝑡} #{𝑏𝑚𝑚 𝑞𝑏𝑢ℎ𝑡}

Results are similar across 36 monitors

14

(60,70)

slide-15
SLIDE 15

Outline

 Background  Motivation  Methodology  Evaluation  Conclusion

15

slide-16
SLIDE 16

Basic idea

16

 aggregate P1 and P2 into P12  Conclusion:  2 fewer changes  4 fewer changes if P3 is further involved

time P1 P2 P3 P2 P1 P6 P7 t1 t2 t3 t4 t5 t6 t7 time P12 P3 P12 P6 P7 t1 t2 t3 t4 t5 t6 t7

slide-17
SLIDE 17

Routing issues

 AS_PATH functions

 Prevent routing loops, influent BGP decision process

 Backup path

 AS 6 and AS 7 are p2p

17

AS 1 AS 2 AS 3 AS 4 AS 5 AS 6 AS 7 AS 8 prefix d

slide-18
SLIDE 18

Solution

 Per peer and per prefix  SSLD(Sender sider loop detection) [Labovitz02]  Example

To AS 8: [7 4 2 1], [7 5 2 1], [7 6 3 1]

 Aggregated path is 7{2 3 4 5 }1

To AS 6: [7 4 2 1], [7 5 2 1]

 Aggregated path is 7{4 5}2 1

18

slide-19
SLIDE 19

Workflow

 Upon receiving a route r regarding p

Update the prefix penalty associated to p Update the path penalty associated to r.path Update the path penalty in p’s history cache If prefix penalty regarding p > threshold

 AS_PATH aggregation is triggered

 Clean process is scheduled every T hours

 Remove those paths whose path penalties

are small enough

19

slide-20
SLIDE 20

Outline

 Background  Motivation  Methodology  Evaluation  Conclusion

20

slide-21
SLIDE 21

Evaluation(1/3)

 Compared with Path Exploration

Damping(PED) [Huston10]

Perform better than MRAI with 35 PEDI

 Metric: reduced updates, convergence

duration, convergence delay

 Performance: better in 29/36 monitors  Convergence: better in all monitors

21

slide-22
SLIDE 22

Evaluation(2/3)

 Compared with RFD  Metric: reduced updates, involved prefixes  Perform better in 21/36 monitors  Suppress more prefixes

22

slide-23
SLIDE 23

Evaluation(3/3)

 Memory cost  AS_PATH sharing  Only upper bound is evaluated

At most 5,000 more paths per router Higher ASes buffer fewer AS_PATHs

23

slide-24
SLIDE 24

Outline

 Background  Motivation  Methodology  Evaluation  Conclusion

24

slide-25
SLIDE 25

Conclusion

 BGP churn is a problem, especially for

those highly active prefixes

 To utilize path locality is a potential choice  Next step is to extend our approach to

iBGP so that AS itself can benefit from this technology as well

25

slide-26
SLIDE 26

26