A Measurement Study of Internet Bottlenecks

1
A Measurement Study of Internet Bottlenecks

• Ningning Hu (CMU)
• Joint work with
• Li Erran Li (Bell Lab)
• Zhuoqing Morley Mao (U. Mich)
• Peter Steenkiste (CMU)
• Jia Wang (ATT)

2
Motivation

• Recent research progress on active probing makes
  it possible to locate bandwidth bottlenecks
• How persistent are the Internet bottlenecks?
• Important for measurement frequency
• What relationship exists between bottleneck and
  packet loss and queuing delay?
• Useful for congestion identification
• Are bottlenecks shared by end users within the
  same prefix?
• Useful for path bandwidth inference
• What causes intra-AS bottlenecks?
• Important for traffic engineering

3
Pathneck

• Bottleneck
• Bottleneck Link the link with the smallest
  available bandwidth on a network path
• Bottleneck Router the downstream router of a
  bottleneck link
• Pathneck
• An active probing tool that can detect Internet
  bottleneck location effectively and efficiently
• For details, please refer to
• Locating Internet Bottlenecks Algorithms,
  Measurements, and Implications SIGCOMM04
• Source code www.cs.cmu.edu/hnn/pathneck
• Pathneck output used in this work
• Bottleneck link location
• Route

4
Data collection
D
D
960 Internet Destinations
D
cmu
S
D
D
D
D

• Probing
• Source a CMU host
• Destinations 960 diverse IP addresses
• 10 continuous probings for each destination (1.5
  minutes)
• Repeat for 38 days (for persistence study)
• Limitations
• Pathneck can not cover the last hop
• 960 ltlt of Internet paths

5
Outline

• How persistent are the Internet bottlenecks?
• Route persistence
• Bottleneck persistence
• What relationship is between bottleneck and
  packet loss and queuing delay?
• Are bottlenecks shared by end users within the
  same prefix?
• What causes intra-AS bottlenecks?

6
Terminology
probing set (persistent)
Day 1
not persistent

• Consider both AS-level route and location-level
  route

7
Route Persistence
AS level
Location level

• Route change is very common and must be
  considered for bottleneck persistence analysis
• Consistent with the results from Zhang, et. al.
  IMW-01 on route persistence

8
Bottleneck Persistence

• Persistence of a bottleneck router
• Bottleneck Persistence of a path
• Max(Persist(R)) for all bottleneck router R
• Two views
• End-to-end view ? per (src, dst) pair
• Includes the impact of route change
• Route-based view ? per route
• Removes the impact of route change

of persistent probing sets R is bottleneck
Persist(R)
of persistent probing sets R appears
9
Bottleneck Persistence
3

1. Bottleneck persistence in route-based view is
   higher than end-to-end view
2. AS-level bottleneck persistence is very similar
   to that from location level
3. 20 bottlenecks have perfect persistence in
   end-to-end view, and 30 for route-based view

10
Outline

1. How persistent are the Internet bottlenecks?
2. What relationship exists between bottleneck and
   packet loss and queuing delay?
3. Are bottlenecks shared by end users within the
   same prefix?
4. What causes intra-AS bottlenecks?

11
Motivation

• Possible congestion indication
• Large queuing delay
• Packet loss
• Bottleneck
• They do not always occur together
• Packet scheduling algorithm ? large queuing delay
• Traffic burstiness or RED ? packet loss
• Small link capacity ? bottleneck
• Bottleneck ??? link loss large link delay

12
Method

• Collected on the same set of 960 paths, but
  independent measurements
• Detect bottleneck location using Pathneck
• Detect loss location using Tulip
• Only use the forward path results
• Detect link queuing delay using Tulip
• medianRTT minRTT
• Tulip was developed in University of
  Washington, SOSP03
• Our analysis is based on the 382 paths for which
  both bottleneck location and packet loss are
  detected

13
Bottleneck ?? Packet Loss
14
Bottleneck ?? Link Delay
15
More Results

1. How persistent are the Internet bottlenecks?
2. What relationship exists between bottleneck and
   packet loss and queuing delay?
3. Are bottlenecks shared by end users within the
   same prefix?
4. What causes intra-AS bottlenecks?

16
Related Work

• Persistence of Internet path properties
• Zhang IMW-01, Paxson TR-2000, Labovitz
  TON-1998, Infocom-1999
• Congestion points sharing
• Katabi TR-2001, Rubenstein Sigmetrics-2000
• Correlation among Internet path properties
• Paxson 1996
• Correlation between router and link properties
• Agarwal PAM 2004

17
Conclusion

• Only 20-30 Internet bottlenecks have perfect
  persistence
• Application should be ready for bottleneck
  location change
• Bottleneck locations have a fairly strong (60)
  correlation with packet loss locations
• Bottleneck and loss detections should be used
  together for congestion detection
• End users within common cluster share bottlenecks
  only with a low probabilityh
• End user can not assume common bottlenecks
• We observe evidence of a correlation between
  bottleneck and link loads
• Network engineers should focus on traffic load to
  eliminate bottlenecks