I-3 content-centric networking Taekyoung Kwon (TK) [email protected] Some slides are from Van...

44
i-3 content-centric networking Taekyoung Kwon (TK) [email protected] Some slides are from Van Jacobson@PARC 1

Transcript of I-3 content-centric networking Taekyoung Kwon (TK) [email protected] Some slides are from Van...

Page 1: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

1

i-3 content-centric networking

Taekyoung Kwon (TK)[email protected]

Some slides are from Van Jacobson@PARC

Page 2: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

2

Why Content Networking (CN)?

Page 3: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

3

Internet• Original motivation for Internet is to share

computing resources– Remote login, file transfer

• The Internet communication model is host-to-host conversations– An IP address indicates a host

• It has been successful for 40 years– TCP/IP has been so flexible for diverse host

based applications

Page 4: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

4

Why content networking (CN)?

• Internet traffic is already content-oriented– CDN, data center, web cache, redundancy elimination… – multimedia, web, P2P, IPTV…

• Users/applications care “what to receive”– They don’t care “from where or from whom”– So, host-based comm. model is outdated

• Storage cost is getting cheaper sharply– Compared to networking cost

• Other advantages of CN coming up later

Page 5: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

5

Internet traffic breakdown

Page 6: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

6

IP networking• Host-centric design• Lookup-by-name

– Indirection (from name to locator)• DNS

– Host/link/DNS availability concern

• DNS is extendible and highly available– Distributed design– Thanks to caching

• Locators can be aggregated– Network prefix

• Currently almost 450k

– Routing scalability is better than CN

Page 7: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

7

Content networking (CN)

• Route-by-name– No indirection, better availability– Content name (or ID) is a routing entry– Huge scalability concern

• In-network caching

• Global-scale pure CN may not be feasible– At least billions of contents– Some aggregation may be possible

• E.g. hierarchical names like URLs

• Other merits such as authentication

Page 8: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

8

Storage cost• In-network storage

Page 9: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

9

IP networking vs. CN

• Network prefix • Content name

Destination Next Hop

192.168.0.0/16 Router C

Content Name Next Hop

/a.com/b.jpg Router C

/a.com/b.jpg

B: /a.com/

b.jpg

Page 10: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

10

Content name or ID

• Content names (Cnames)– May replace the IP addresses– Content identifiers (CIDs)

• Cname/CID design choices– Hierarchical vs. flat – Semantics vs. semantic-free

• Persistency• Location independence

– Variable length vs. fixed length

• examples– cnn.com/sports/news.avi– sonypictures.com/spiderman3.html– yahoo.co.kr/image/logo.jpg– 0xF034BC….024A,

• E.g. hash of content data, name, public key

– Or hybrid

Page 11: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

11

Advantages of CN

• Better delivery efficiency– Multicast, mobility, QoS/QoE,…

• Caching at CN-capable routers– Shortest path to the (potentially cached) content

• Inter-ISP traffic reduction• Web host provisioning

– Server-less computing

• Policing– track the history of content requests

• Content authenticity– signature– Phishing and pharming are not possible

Page 12: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

12

Where to put Cname?

• In TCP/IP– Application layer header

• E.g. HTTP, SIP• Deep packet inspection

– IP option header

• New L3 header– A clean slate approach

Page 13: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

13

Content Centric Networking (CCN)Named Data Networking (NDN)

Page 14: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

14

CCN

• “Networking Named Content,” ACM CoNEXT 2009.

• Van Jacobson – Palo Alto Research Center (PARC)

• NDN project at NSF

Page 15: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

15

CCN Philosophy

• Solve the cognitive mismatch– User/app wants “what”– Network wants “who”– Mapping between two models requires a lot of

convention and configuration (middleware, wet-ware)

• Users specify the objective• No distinction between bits in a memory and

in a wire• Data security and integrity are the architec-

tural foundation

Page 16: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

16

CCN basics• Content name

– Hierarchical, variable-length, semantics

• No IP address

• Consumers send Interest Packets• Content holders send back Data Packets

Source: Van Jacobson@PARC

Page 17: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

17

CCN hourglass model

Source: Van Jacobson@PARC

Page 18: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

18

A user wants some content

Source: Van Jacobson@PARC

Page 19: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

19

Content is downloaded

Content is cached!

In-network caching

Source: Van Jacobson@PARC

Page 20: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

20

Another user requests the same content

Source: Van Jacobson@PARC

Page 21: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

21

Name tree

• Components• Tree traversal to discover

data• Default traversal is Left-

mostChild• Relations: prev, next• /parc.com/videos/Wid-

getA.mpg RightmostChild Source: Van Jacobson@PARC

Page 22: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

22Source: Van Jacobson@PARC

CCN forwarding

Page 23: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

23

CCN: Strengths and Weaknesses

Pros• Better availability• Better delivery• Inter-ISP traffic• Accountability, Policing• Integrity• Authentication• Serverless computing• Multicast• Mobility

Cons• Huge scalability con-

cern• More in-network pro-

cessing• Potentially more sig-

naling for routing

Page 24: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

24

Multiple copies

CCN router

IP router

* No loop in CCN!

Source: Van Jacobson@PARC

Page 25: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

25

CCN security

• In CCN, the content itself (not its container) is trusted– In TCP/IP, endpoints are authenticated

• Anypoint can retrieve the content from anywhere and validate it– Content is publicly authenticatable

• All content is digitally signed– Binding btw. name and the content itself

• Still PKI is needed

Page 26: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

Signature in CCN• Binding btw content name and payload is au-

thenticated

26

Publisher’s private key

Publisher’s public key

digitalSignature(encrypt)

Name || Pay-load

publisher

Name ||Payload ||Signature

Data packet

signature

anypoint

digitalSignature(decrypt)

Name ||Payload

PKI

Verify!

Page 27: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

27

Interest packet

Source: Udugama at Univ. Bremen

Page 28: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

28

Data packet

Source: Udugama at Univ. Bremen

Page 29: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

29

CCN vs. Related Technologies(CDN, P2P, ICN)

Page 30: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

30

Why CDN? For CPs!

• How do I get my content to my cus-tomer quickly, reliably, and accu-rately?

• How can I support 20 million hits per day?– Flash crowd

• Can I offload any server traffic?

* CP: content providerCDN: content delivery net-work

Page 31: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

31

CDNs make content more available

• Push the content to the edge– Multiple places

• Load balance mirrored content• Creative DNS solutions

Page 32: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

32

First Point – DNS

• Selects from among several mirror sites operated by content provider

Page 33: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

33

End User

Akamai DNS Resolution

Akamai High-Level DNS Servers

10g.akamai.net

1

Browser’s Cache

OS

2

Local Name Server3

xyz.com’s nameserver

6ak.xyz.com

7a212.g.akamai.net

915.15.125.6

16

15

1120.20.123.55

Akamai Low-Level DNS Servers

12 a212.g.akamai.net

30.30.123.513

14

4xyz.com .com .net

Root(Verisign)

10.10.123.55

akamai.net

8

select cluster

select servers within cluster

Page 34: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

34

With/without Akamai

Akamai operates over 73,000 servers in 70 countries in about 1,000 auton-omous systems, which on any given day may handle upwards of 20% of traffic in 2010

Page 35: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

35

Akamai vs. LimeLight (As of 2008)

Source: GridsLab at Univ. of Melbourne

Page 36: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

36

Problems with legacy CDNs

• Current CDN models good for offloading content distribution

• Cache mostly lo-cated at exchange points– Colocation

• No benefit to access network operators– No traffic reduction– No revenue sharing Source: Dirk Kutcher@NEC Lab

Page 37: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

37

telco CDN (or operator CDN)

• ISPs are not happy with “off-net” CDNs– Network control issues– Limited monetization chances

Source: Alcatel-Lucent

Page 38: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

38

telco CDN: strengths

• Aka “On-net” CDN• Optimize QoS/QoE

– E.g. Deep caching

• Low cost– Same CDN infra for its own content and CP’s content– Cache OTT content (i.e. reduce traffic cost)

• Help CPs– Customize services depending on content portfolio

• New Biz models– In the content-to-customer chain

• Managed vs. OTT content• E.g. Value-added service for CPs* OTT: over the top

Page 39: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

39

Recent changes in CDN: P2P CDNs

• P2P CDNs– Exploit user machines mostly– Little cost– Often copyright issues

Page 40: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

40

Google Global Cache

• CP can be a CDN provider as well

Page 41: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

www.bittorrent.com

Peer

1

popeye.mp4.torrent

• File popeye.mp4.torrent hosted at a (well-known) webserver

• The .torrent has ad-dress of tracker for file

• The tracker, which runs on a webserver as well, keeps track of all peers downloading file

BitTorrent operations for “popeye.mp4”

41

Page 42: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

Peer

TrackerAddresses of peers

2

www.bittorrent.com

• File popeye.mp4.torrent hosted at a (well-known) webserver

• The .torrent has ad-dress of tracker for file

• The tracker, which runs on a webserver as well, keeps track of all peers downloading file

BitTorrent operations for “popeye.mp4”

42

Page 43: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

Peer

Tracker3

www.bittorrent.com

Swarm

• File popeye.mp4.torrent hosted at a (well-known) webserver

• The .torrent has ad-dress of tracker for file

• The tracker, which runs on a webserver as well, keeps track of all peers downloading file

BitTorrent operations for “popeye.mp4”

43

Page 44: I-3 content-centric networking Taekyoung Kwon (TK) tkkwon@snu.ac.kr Some slides are from Van Jacobson@PARC 1.

44

BitTorrent: Basic Idea

• Chop a file into many pieces• Replicate DIFFERENT pieces on dif-

ferent peers as soon as possible• As soon as a peer has a complete

piece, it can trade it with other peers– Tit-for-tat

• Hopefully, a peer will be able to as-semble the entire file at the end