Recent Advances in Network Computing

Zhiwei Zhao
zzw@uestc.edu.cn


mobinets @ SCSE, UESTC
1. Named Data Network 2. Turing Awards 3. Edge Computing

Two lines

Recent advances Turing award winners
1. NDN/SDN/NFV 1. Stories
2. Edge computing 2. Interesting works
3. Lectures

Coursework (60%+40%)

A one-column, two-page course work.

Sample pdf, Template tex file.

Attendance

The roll will be called when many are absent.
No final score if you are found absent for ≥3 times.

Submission

Email: fmi_course@mobinets.org
Deadline: May 26, 2022.

Recent advances

NDN: Named Data Networking.

  • An instance of ICN[1][2].
  • Data has names and can be identified.

SDN: Software Defined Networks

NFV: Network Function Virtualization

Edge: Edge Computing

  • Cloud in proximity.
  • Dispersed and resource constrained.

Named Data Networks

Addressing: IP vs ID

[1] Lixia Zhang et al. Named data newtorking, ACM SigComm 2014.

Software Defined Network

[1] N Gude et al. NOX: Towards an Operating System for Networks. ACM SigComm 2008.
[2] N McKeown et al. OpenFlow: Enabling Innovation in Campus Networks. ACM SigComm 2008.

Network Function Virtualization

[1] ESTI, Network Functions Virtualisation, White Paper.
[2] Bo Han et al. Network function virtualization: Challenges and opportunities for innovations, IEEE Comm. Mag. 2015.

Edge Computing

[1] Mahadev Satya et al. The Case for VM-based Cloudlets in Mobile Computing, IEEE Pervasive Computing, 2009.
[2] Mahadev Satya, The Emergence of Edge Computing, Computer, 2017.
[3] J Chen and X Ran, Deep learning with edge computing: A review, Proc. of IEEE, 2019.

👑 Turing Award

The ACM Alan Mathison Turing Award - "Nobel Prize of Computing".

56 Turing Awards

Field Awards Percentage
Programming lauguage 16 28.6%
Computing theory 12 21.4%
Network system 8 14.3%
Database 7 12.5%
AI 6 10.7%
CHI 4 7%
Security 3 5.4%

Stars in networks (1)


Ken Thompson & Dennis Ritchie (1983)
Richard M. Karp (1985)

Fernando J. Corbató (1990)
Butler W. Lampson (1992)

Stars in networks (2)


Vint Cerf & Robert Kahn (2004)
Charles P. Thacker (2009)

Leslie Lamport (2013)
Tim Berners-Lee (2016)

On our stage..


Vint Cerf & Robert Kahn (2004)
Richard M. Karp (1985)
Leslie Lamport (2013)

🤔
Explore their problems
📜
Seminal works
🎤
Stories and Interviews

Network ≈ Blackhole!

More and more apps become networked

Overleaf, cloud games, cloud native apps, etc.

Reason?

Human desires: Service close, cost away

Driver: Evolutional Internet

Inter...net?

Inter意为 ...之间

Interaction 互动、交流
International 国际的、跨国的
Internet 跨...网的？？

Internet's essential goals



Addressing & Switching


* Internet and Telecommunication network？

Interconnected networks

网络类型 带宽分配Mbps 占比%
中国电信 4,537,680 50.72
中国联通 2,234,738 24.98
中国移动 1,997,000 22.32
中国科技网 115,712 1.29
CERNET中国教育
和科研计算机网		 61,440 0.69
总计 8,946,570 100

中国科技网

Inet vs Tnet

  • Internet wishes: Inter-connections (IETF, IEEE)
  • Tele. Net wishes: Business (ITU, 3GPP)
  • Relationships
    • Tnet is part of Inet
    • Inet wants more networks
    • Tnet wants more users


* 全球与漂亮国？

Advances in Internet

📌 Addressing

  • Content centric network
  • Named Data Network


🍻 Switching

  • Quality of Service
  • Quality of Experience
  • End-to-end requirements

News in various networks

  • 3G/4G/5G/6G
  • Orbital
  • Visible light
  • Low-power

Novel Information and Communication Theories, and new networking technologies driven by them.

Telecommunications

Name Bandwidth Coverage
3G 2Mbps 2-5km
4G 100Mbps 1-3km
5G 1Gbps 50-300m
6G 1Tbps 10m

Bandwidth↗ Coverage↘ Mobility↗

Apps are dragged into network as the end-to-end latency decreases and throughput increases.

Network computing

Centralized🔁Distributed

* 刘云新. 智能边缘计算中的挑战和机遇.中国计算机学会通讯 2021.09. * 汪学海等. 边缘计算技术研究报告. 中科院计算所 信息技术与信息化前瞻FIAR-02. 2018.02.

Grow ripe

Tech. vision

Embedding each other

  • C → N: SDN、NFV
  • N → C: Cloud、Edge、D2D/D4D

Edge computing

边缘计算是一种计算资源与用户接近、计算过程与用户协同、整体计算性能高于本地和云计算的计算模式，是实现无处不在的“泛在算力”的具体手段。

* 机械工业出版社《边缘计算:原理、技术与实践》（赵志为、闵革勇编著）

Needs

Internet-of-Things

More "things" are getting smart, with increasing computational needs.

Internet-of-Things

IoT Brief History

📥 Build a network for things

🔗 Connect things into a network

🌃 Connected everything

IoT Essense

IoT-Human gap mainly lies in sensing and computing.

AI+IoT=AIoT

Advances in network brings computing on board.

🏡IoT 🔄Link 🖥Server
Offloading decisions Task transfer Task processing
Analysis/divisions Multi-access Service management

A brief history of edge

  • 2003:EdgeComputing (Akamai&IBM).
    •
  • 20th: Cloud rapidly popularized.
    •
  • 2008: Cloudlet was clearly put forward.
    •
  • 2012: Concept of Fog Computing.
    •
  • 2013: ESTI defines Mobile Edge Computing.
    •
  • 2015: Open Fog Computing Alliance.
    •
  • 2016: ACM/IEEE Symp. on Edge computing.
    •
  • 2017: Edge Computing Industry Alliance.
    •
  • 2018: MEC re-explained.
    •
  • 2019: Edge/Green Computing Commitee.
    •
  • 2020: Scene-oriented edge systems (OEC).
    •

Gartner's view

Year Keywords Stage
2017 Edge Computing Innovation Trigger
2018 Edge AI Innovation Trigger
2019 Edge AI/Analytics Peak of inflated expectations
2020 Low-cost single-board computers at edge Innovation Trigger

* TinyML, Embedded AI, Lightweight ML, etc.
 * Gartner's Hype Cycle.

EdgeAI or CloudAI?

Distinguishment: Scenarios matter.

  • Access & Offloading: Ambient and unreliable.
  • Deployment: High cost
  • Architecture: Cloud-edge-end

* Integrating sensing, communication and computation.

AI everywhere?

Mobility is perhaps the first challenge.

Scene Speed Handovers
AutoDrive+5G 60 KM/h 5 AP/m
HighRail+5G 300 KM/h 25 AP/m
AR/VR+5G 4 KM/h 0.33 AP/m
Wander+WiFI6 4 KM/h 0.33 AP/m

🧐Where are the impacts?
service handover, cooperation, sys deployment, service discovery, ...

Explosive app growth

Almost all services can be trasferred to the edge


  • Resource: constrained
  • Types:Heterogeneous
  • Service amount:Huge
Massive system works to be done......

AIoT/Edge Papers

  • SigComm'18, AuTO: Scaling Deep Reinforcement Learning for Datacenter-Scale Automatic Traffic Optimization
  • SOSR'21, Accelerating Distributed Deep Learning using Multi-Path RDMA in Data Center Networks
  • SigComm'21, MimicNet: Fast Performance Estimates for Data Center Networks with Machine Learning
  • SigComm'20, Interpreting Deep Learning-Based Networking Systems

Some more

  • MobiCom'21, EMP: Edge-assisted Multi-vehicle Perception
  • MobiCom'21, Flexible High-resolution Object Detection on Edge Devices with Tunable Latency
  • MobiCom'19, Edge Assisted Real-time Object Detection for Mobile Augmented Reality

Interesting videos🚴‍♀️

AI + Wireless + Mobile (1)

FL: similarity aware cluster for activity recognition.
remote gesture: real time graph analysis & control.
mole: sensing input with smartwatch.
AutoDrive1, AutoDrive2, AutoDrive3.
gesture1, gesture2, gesture3: sensing gestures.
mobile AR: mobile augmented reality (early).
Cable robot, userview: 360 VR.
mblocks: self learning, sensing and navigating.
pocket edge: low cost edge server using smartphones.
UAV network: Self organizing and target tracking.

AI + Wireless + Mobile (2)

Apple Adhoc: Apple wireless direct link.
Throughput fairness: Tradeoffs in mobility platforms.
Underwater VLC: novel communications.
UAV side channel: Audio side-channel for UAVs.
3D human mesh, Crowd counting: sensing wirelessly
IoT democratization: hmm.. interesting topic.
Find missing objects: hybrid use of vision and sensing.
Liquor detect: sensing with graphics.