Drafting Behind Akamai (Travelocity-Based Detouring)

24
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/~ak uzma/

description

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/. Drafting. Detour. Drafting Behind Akamai. Motivation. - PowerPoint PPT Presentation

Transcript of Drafting Behind Akamai (Travelocity-Based Detouring)

Page 1: Drafting Behind Akamai (Travelocity-Based Detouring)

Drafting Behind Akamai (Travelocity-Based Detouring)

Aleksandar KuzmanovicNorthwestern University

Joint work with:A. Su, D. Choffnes, and F. Bustamante

To appear in Sigcomm 2006

http://www.cs.northwestern.edu/~akuzma/

Page 2: Drafting Behind Akamai (Travelocity-Based Detouring)

2 A. Kuzmanovic Drafting Behind Akamai

Drafting Behind Akamai

Drafting

Detour

Page 3: Drafting Behind Akamai (Travelocity-Based Detouring)

3 A. Kuzmanovic Drafting Behind Akamai

Motivation

Overlay networks– Can’t 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

Page 4: Drafting Behind Akamai (Travelocity-Based Detouring)

4 A. Kuzmanovic Drafting Behind Akamai

ProblemIndependent 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

Page 5: Drafting Behind Akamai (Travelocity-Based Detouring)

5 A. Kuzmanovic Drafting Behind Akamai

Our ApproachFact– 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

Page 6: Drafting Behind Akamai (Travelocity-Based Detouring)

6 A. Kuzmanovic Drafting Behind Akamai

CDN-Driven One-Hop Source Routing

S

A1

An

A2

……..

D

DNS Server

En

E2

E1

Page 7: Drafting Behind Akamai (Travelocity-Based Detouring)

7 A. Kuzmanovic Drafting Behind Akamai

Key QuestionsHow 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

Page 8: Drafting Behind Akamai (Travelocity-Based Detouring)

8 A. Kuzmanovic Drafting Behind Akamai

DNS “Black Magic”

WebClient

Akamai High-LevelDNS Server

Akamai Low-LevelDNS Server

Local DNS ServerPCWorld AuthoritativeDNS Server

PCWorld Web ServerAkamai

Edge Server

Root DNS Server

Top-Level Domain DNS Server

www.pcworld.com

www.pcworld.comwww.pcworld.comwww.pcworld.com

http request/response

images.pcworld.com

CNAME: a1694.g.akamai.netimages.pcworld.com a1694.g.akmai.net

a1694.g.akmai.net2 ip addresses of Akamai Edge Servers2 ip addresses of Akamai Edge Servers

fetch image files

Page 9: Drafting Behind Akamai (Travelocity-Based Detouring)

9 A. Kuzmanovic Drafting Behind Akamai

Measuring Akamai2-months long measurement140 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.

PL Node

AkamaiEdge Server 2

AkamaiEdge Server 3

Akamai Low-LevelDNS Server .……

AkamaiEdge Server 1

Page 10: Drafting Behind Akamai (Travelocity-Based Detouring)

10 A. Kuzmanovic Drafting Behind Akamai

Initial Results

day

night

Berkeley Purdue

Page 11: Drafting Behind Akamai (Travelocity-Based Detouring)

11 A. Kuzmanovic Drafting Behind Akamai

Server Diversity for Yahoo

Good overlay-to-CDN mapping candidates

Majority of PL nodessee between 10 and 50 Akamai edge-servers

Nodes far away from Akamaihot-spots

Page 12: Drafting Behind Akamai (Travelocity-Based Detouring)

12 A. Kuzmanovic Drafting Behind Akamai

Multiple Akamai Customers

Amazon is not hosted on UMass and LBNL

Yahoo and NYTimes are not hosted in U. Oregon

FEMA is poor!

Page 13: Drafting Behind Akamai (Travelocity-Based Detouring)

13 A. Kuzmanovic Drafting Behind Akamai

Redirection Dynamics

Berkeley Brazil

Korea

Page 14: Drafting Behind Akamai (Travelocity-Based Detouring)

14 A. Kuzmanovic Drafting Behind Akamai

Key QuestionsHow 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

Page 15: Drafting Behind Akamai (Travelocity-Based Detouring)

15 A. Kuzmanovic Drafting Behind Akamai

Methodology

Planet Lab NodeAkamai Low-Level

DNS Server

10 Best Akamai

Edge Servers

………

Page 16: Drafting Behind Akamai (Travelocity-Based Detouring)

16 A. Kuzmanovic Drafting Behind Akamai

Do CDN redirections correlate to network latencies? (1)

Rank = r1+r2-1– 16 means perfect correlation – 0 means poor correlation

Brazil is poor

MIT and Amsterdam are excellent

Page 17: Drafting Behind Akamai (Travelocity-Based Detouring)

17 A. Kuzmanovic Drafting Behind Akamai

Do CDN redirections correlate to network latencies? (2)

CNN is poor!

70% paths are among best 10%

97% better than average

Page 18: Drafting Behind Akamai (Travelocity-Based Detouring)

18 A. Kuzmanovic Drafting Behind Akamai

Akamai-Driven One-Hop Source Routing

S

A1

An

A2

……..

D

DNS Server

Redirection dynamics sufficiently small for

network control

En

E2

E1

Redirections driven by network conditions

Potential for CDN-to-overlay mapping

Page 19: Drafting Behind Akamai (Travelocity-Based Detouring)

19 A. Kuzmanovic Drafting Behind Akamai

Methodology

Source

Akamai Low-LevelDNS Server

Destination

Direct PathAkamai path 3

Akamai path 1

…….....

Akamai path 10

Akamai path 2

Akamai path 1

Akamai path 2

Akamai path 3

Akamai path 10

Page 20: Drafting Behind Akamai (Travelocity-Based Detouring)

20 A. Kuzmanovic Drafting Behind Akamai

Akamai-driven source routing (1)

80% Taiwan15% Japan5 % U.S.

75% U.K.25% U.S.

Taiwan-UK UK-Taiwan

Page 21: Drafting Behind Akamai (Travelocity-Based Detouring)

21 A. Kuzmanovic Drafting Behind Akamai

Akamai-driven source routing (2)

Direct paths betterthan Akamai paths

25% of Akamaipaths better thandirect pathsAkamai and direct

paths have similar performance

Experiment: US (6), Europe (3), S. America (2), Asia (3)

Page 22: Drafting Behind Akamai (Travelocity-Based Detouring)

22 A. Kuzmanovic Drafting Behind Akamai

Path pruningFact– 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)

Page 23: Drafting Behind Akamai (Travelocity-Based Detouring)

23 A. Kuzmanovic Drafting Behind Akamai

Path Pruning Result

Direct path accounts for

78% of the gain

2 hours update frequency before the performance

declines

BTAS better than FAS

Page 24: Drafting Behind Akamai (Travelocity-Based Detouring)

24 A. Kuzmanovic Drafting Behind Akamai

ConclusionsReuse measurements performed by CDNsReverse-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