Drafting Behind Akamai TravelocityBased Detouring

1
Drafting Behind Akamai (Travelocity-Based
Detouring)

• Aleksandar Kuzmanovic
• Northwestern University
• Joint work with
• A. Su, D. Choffnes, and F. Bustamante
• To appear in Sigcomm 2006

http//www.cs.northwestern.edu/akuzma/
2
Drafting Behind Akamai
3
Motivation

• Overlay networks
• Cant change IP, so change the layers above
• E.g., end-point multicast
• Common problem
• Build the view of the underlying network
• Passive and active measurements

4
Problem

• Independent measurements
• Redundant
• Non-scalable
• Can cause problems
• E.g., synchronization
• Common service needed
• Knowledge plane - MIT
• A routing underlay for overlays - Princeton
• Network Weather Service - Berkeley

5
Our Approach

• Fact
• CDNs (e.g., Akamai) perform extensive network and
  server measurements
• Publish the results through DNS over short time
  scales
• Can overlay networks reuse measurements collected
  by production CDNs?
• Significantly reduce the amount of measurements
  (a complementary service)
• No new infrastructure need to be deployed
• Inherit the robustness of DNS
• Easy integration with existing systems

6
CDN-Driven One-Hop Source Routing
7
Key Questions

• How does Akamai work?
• DNS translation
• How many web replicas does a client see?
• Impact of different sites (e.g., Yahoo vs. NY
  Times)?
• DNS redirection dynamics?
• Network or server latency?
• An example application one-hop source routing
• Mapping CDN servers to overlay nodes
• Low-overhead protocols for exploiting CDN
  redirections

8
DNS Black Magic
www.pcworld.com
www.pcworld.com
a1694.g.akmai.net
CNAME a1694.g.akamai.net
images.pcworld.com
www.pcworld.com
a1694.g.akmai.net
2 ip addresses of Akamai Edge Servers
2 ip addresses of Akamai Edge Servers
www.pcworld.com
images.pcworld.com
fetch image files
http request/response
9
Measuring Akamai

• 2-months long measurement
• 140 PlanetLab (PL) nodes
• 50 US and Canada, 35 Europe, 18 Asia, 8 South
  America, the rest randomly scattered
• Every 20 sec, each PL node queries an appropriate
  CNAME for
• Yahoo, CNN, Fox News, NY Times, etc.

.
10
Initial Results
Berkeley
Purdue
11
Server Diversity for Yahoo
Majority of PL nodes see between 10 and 50
Akamai edge-servers
12
Multiple Akamai Customers
13
Redirection Dynamics
14
Key Questions

• How does Akamai work?
• DNS translation
• How many web replicas does a client see?
• Impact of different sites (e.g., Yahoo vs. NY
  Times)?
• DNS redirection dynamics?
• Network or server latency?
• Potentials for one-hop source routing?
• Mapping CDN servers to overlay nodes
• Low-overhead protocols for exploiting CDN
  redirections

15
Methodology
10 Best Akamai Edge Servers

16
Do CDN redirections correlate to network
latencies? (1)

• Rank r1r2-1
• 16 means perfect correlation
• 0 means poor correlation

17
Do CDN redirections correlate to network
latencies? (2)
18
Akamai-Driven One-Hop Source Routing
Redirections driven by network conditions
Potential for CDN-to-overlay mapping
Redirection dynamics sufficiently small for
network control
19
Methodology
20
Akamai-driven source routing (1)
Taiwan-UK
UK-Taiwan
21
Akamai-driven source routing (2)
Experiment US (6), Europe (3), S. America (2),
Asia (3)
22
Path pruning

• Fact
• Not always is Akamai-driven path better than the
  direct one
• Practical issues
• How frequently to make a decision whether to use
  the direct or the Akamai path?
• Should one use
• the first (of the 2 paths) returned by Akamai
  (FAS)
• the better (of the 2 paths) returned by Akamai
  (BTAS)

23
Path Pruning Result
24
Conclusions

• Reuse measurements performed by CDNs
• Reverse-engineering Akamai
• DNS redirections sufficiently small
• Strong correlation to network conditions
• All clients see a large number of paths
• CDN-driven one-hop source routing
• 25 of Akamai paths outperform direct paths
• 50 of nodes discovered by Akamai outperform
  direct paths
• Low-overhead pruning algorithms
• Global Internet weather-report service for
  little to no cost