11
1 Network science (NS): hype or reality? Chuanxiong Guo Microsoft Research Asia September 24, 2009

1 Network science (NS): hype or reality? Chuanxiong Guo Microsoft Research Asia September 24, 2009

  • View
    214

  • Download
    0

Embed Size (px)

Citation preview

Page 1: 1 Network science (NS): hype or reality? Chuanxiong Guo Microsoft Research Asia September 24, 2009

1

Network science (NS): hype or reality?

Chuanxiong GuoMicrosoft Research Asia

September 24, 2009

Page 2: 1 Network science (NS): hype or reality? Chuanxiong Guo Microsoft Research Asia September 24, 2009

2

Definition

• Science (Wikipedia)– Any systematic knowledge-base or prescriptive

practice that is capable of resulting in a prediction or predictable type of outcome

• Network Science (the “Network Science” book)– The study of network representations of physical,

biological, and social phenomena leading to predictive models of these phenomena

Page 3: 1 Network science (NS): hype or reality? Chuanxiong Guo Microsoft Research Asia September 24, 2009

3

Predictability• Internet collapse in 1980s

– TCP congestion control• Network expansion

– 32-bit IPv4 address and 16-bit AS number– Will IPv6 takeover?

• Network security – DDos, virus, worm, and spam

• Network traffic and topology– Self similarity, power law

• Network applications– Web

Page 4: 1 Network science (NS): hype or reality? Chuanxiong Guo Microsoft Research Asia September 24, 2009

4

Networking research: An engineering perspective

• Classical networking topics– Network architecture/protocol/applications– Multiple access– Packet scheduling/switching/routing– congestion control, traffic engineering/measurement,

resource management• New technology trends, user requirement,

economics of scale– P2p, sensor networks, network security, data center

networking, social networking

Page 5: 1 Network science (NS): hype or reality? Chuanxiong Guo Microsoft Research Asia September 24, 2009

5

The emerging of network science

• Inter-disciplinary – Physicist, mathematician, sociologist, information

theorist, computer scientist, economist (?)• Data-driven discovery

Page 6: 1 Network science (NS): hype or reality? Chuanxiong Guo Microsoft Research Asia September 24, 2009

6

Data–driven discovery

• Spam and botnet detection– Spamming Botnets [Xie08-sigcomm]– Spamalytics [Kanich08-ccs]

• Network diagnostics, profiling – [Ionut08-sigcomm, Kandula08-sigcomm]

• Online social networking– [Mislove07-IMC, Nazir08-IMC]

Page 7: 1 Network science (NS): hype or reality? Chuanxiong Guo Microsoft Research Asia September 24, 2009

7

Data-driven discovery: system infrastructure

• Build large scale computing infrastructure for researchers from different disciplines – Scalable data center systems and networks (NSF

CLuE program)– Dryad, Hadoop– DryadLinQ, MapReduce

• Data, data, data!– Accumulation and open access

Page 8: 1 Network science (NS): hype or reality? Chuanxiong Guo Microsoft Research Asia September 24, 2009

8

What role can the wireless network and mobile computing community play?

• Mobile data services– Recent advances in smart-phones – Access bandwidth

• Mobility patterns– [Gonzalez08-nature, Lee09-infocom]– [Wang09-ScienceExpress]

• Mobile communication networks– [Palla07-nature, Seshadri08-kdd]

Page 9: 1 Network science (NS): hype or reality? Chuanxiong Guo Microsoft Research Asia September 24, 2009

9

Wang-Sciencexpress09 Understanding the Spreading Patterns of Mobile Phone Viruses

Page 10: 1 Network science (NS): hype or reality? Chuanxiong Guo Microsoft Research Asia September 24, 2009

10

Summary• Networking research currently is mainly an

engineering field– With research problems generated directly from real

world– “We believe in rough consensus and running code“– David

Clark • Bridging engineering and network science

– Data-driven discovery– New models and insights from the broader science

community– More predictable designs

Page 11: 1 Network science (NS): hype or reality? Chuanxiong Guo Microsoft Research Asia September 24, 2009

11

Network Science

Network Engineering

tools

Data (and

problem)

Infrastructure