Riding along with the Time-Traveling Networking Researcher Mostafa - PowerPoint PPT Presentation

Riding along with the Time-Traveling Networking Researcher Mostafa H. Ammar School of Computer Science, Georgia Tech ammar@cc.gatech.edu

The Time Traveling Networking Researcher Ø 1985 Ø 2018

The Time Travelling Networking Researcher Ø 1985 Ø 2019

Then and Now Content Distribution 1985 2019

Then and Now How we did research Ø ACM SIGCOMM Ø ACM SIGCOMM 1985 2019 Ø 22 papers Ø 32 papers Ø Authors/paper = 1.9 Ø Authors/paper =6 Min =1, Max = 3 Min = 1, Max = 17 Ø Sample Title: Ø Sample Title: “Window Selection in “Gentle flow control: Flow Controlled avoiding deadlock in Networks” lossless networks”

Then and Now Research Approach Ø Design, Build and Ø Build, Measure,Tweak, Enjoy Repeat Forever

Then and Now My Research

What did the Journey look like?

Caveats Ø Adding structure and not re-writing history. Ø Hindsight is 20/20 Ø It is difficult to make predictions, especially about the future!

A Story in Three Chapters Ø How to say “Necessity is the Mother of Invention” in Networkingese? Ø From many, one Ø From one, many

Takeaways Ø A fundamental iterative process governs successful evolution in deployed infrastructure. Ø There is nothing fundamental about a single global network delivering all services. Ø New ManyNets world is upon us: § flexibility bodes well for the future of networking and networking research.

“Necessity is the mother of invention” Ø In Network Speak § Necessity == Service Support or Scale § Invention == Infrastructure Deployment

The Service-Infrastructure Cycle New Network Service or Increased Scale Establish/Upgrade Infrastructure To Meet Scale or Provide Service

Motivators Ø Mesh, any-to-any connectivity Ø Scalability

The Service-Infrastructure Cycle Ø Worked for almost 40 years to produce the current Internet Ø Until in 2005 … complaints about “the inability [of the Internet] architecture to adapt to new pressures and requirements."

Ossification

Ossification Explained New Network Service or Increased Scale Establish/Upgrade Infrastructure To Meet Scale or Provide Service Ossification is when cycle gets stuck

But the Cycle is fundamental Ø Case Studies § Unicast Routing § Multicast Routing

The Cycle in Action: Unicast Routing Ø Unicast Routing: How to find and deploy network paths to single destination. Ø As the Internet grew routing became more complicated

Internet Growth Ossification Prosser D. V . Link State EGP BGP

Cycle in evidence Ø BGP was “built on experience gained with EGP as defined in RFC 904 and EGP usage in the NSFNET Backbone as described in RFC 1092 and RFC 1093 "

The Case of Multicast Ø Multicast : is the act of sending a message to multiple receivers using a single local “transmit” operation. Ø It is highly Scalable

Unicast and Multicast Data Copying Multicast Saves

Multicast and Content Distribution Ø Development anticipated the cycle and did not follow it. Ø Not widely deployed today

Multicast anticipated the Cycle Ossification Dalal & Metcalfe PIM SSM Large Aguilar Groups

So … Ø Before Ossification, the Cycle operated to continuously enable novel applications and increased scale. Ø Success when Cycle is followed Ø Cycle is inoperable due to ossification

From Many, One A single global infrastructure consumes all services

Then: Many Networks Ø A network for work ca. 1987 Ø A network for home Ø Ø Tymnet (USA –public) (AOL) Ø GTE Telenet (USA- public) Ø A network for Ø Transpac (France - public) experimentation Ø Datapac (Canada – public) Ø ARPANet (USA – research) Ø BNA (Boroughs) Ø SNA (IBM) Ø DNA (Digital)

The One Network Revolution Ø Having one network for everything is not an obvious choice. Ø Under-appreciated networking milestone Ø Was driven by § Connectivity as the main goal § Need for Scalability § Economies of scale

OneNet effect on networking research Ø Stifled Innovation: Ideas rejected if § Not deployable on Internet § Not Scalable Ø Difficult to validate without disrupting infrastructure § Eternal quest for experimentation environments § (MBone, QBone, 6Bone, PlanetLab, GENI)

OneNet Contributed to Ossification Ø By Definition: ManyNets cannot be ossified

So … If we have OneNet and it is ossified, are we done yet?

From One, Many How the OneNet is fragmenting

The Service-Infrastructure Cycle New Network Service or Increased Scale Establish/Upgrade Infrastructure To Meet Scale or Provide Service

The Unstoppable Force Ø The demands on the network continue to evolve § Increased scale § Content Providers reign supreme § Application-specific requirements, low latency, prioritization

What Happens When … Unstoppable Force Ossified Immovable Object

The Fragmentation of the Internet Ø The Ossified Internet is yielding Ø Are we seeing the return of ManyNets? § Let’s look at the early signs

Is Google using the Internet?

The Content-Delivery Network Ø Shift from Connectivity to Content Ø Content Provider § Needs to maximize quality of experience § Reduce reliance on intermediaries § Control interaction

The Flattening of the Internet Rest of the Internet Access ISP End Users 41

CP Extensions in Access Networks Ø Embedded Content Provider Appliances Ø ISP Interfaces to allow CP Coordination and Collaboration. Ø A large majority of traffic sees a tiny portion of the global Internet.

“Zero-Hop” Networks Rest of the Internet Access ISP End Users

Dedicated (Bypass) networks Ø SpreadNetworks Ø SIGFOX Ø Haste Ø FirstNet Ø Althea

5G Slicing Ø 5G = Future Mobile Broadband and Beyond § Enhanced mobile broadband (eMBB) § Ultra-reliable and low latency communications (URLLC) § Massive machine type communications (mMTC) http://www.huawei.com/minisite/hwmbbf16/insights/5G-Nework-Architecture-Whitepaper-en.pdf

What does the ManyNets world look like? Servers Rest of the Internet Bypass CP Bypass ZeN Community ISP ISP ISP Users

So What? Ø Fragmentation is a reality let’s embrace it. § Attempts to fight ossification without fragmentation failed (see “Clean Slate”) § Fragmentation opens up the research agenda. Ø Loss of OneNet advantages: § Economy of scale § Low barrier to entry

Takeaways Ø A fundamental iterative process governs successful evolution in deployed infrastructure. Ø There is nothing fundamental about a single global network delivering all services. Ø New ManyNets world is upon us: § flexibility bodes well for the future for networking research

Epilogue: The Exciting Future of Networking Ø Scale and new services will continue to drive evolution Ø Ossification will be a thing of the past § The ManyNets world will be a fertile ground for innovation § Network Programmability will add to its responsiveness

Onwards to 2050

The Internet License Office

Internet Licenses Ø Started being required in 2035! Ø Jurisdictions came to the realization that un-regulated highly scalable digital connectivity can be harmful. Ø Solution: Scope Fragmentation and Speed Control.

Internet License Categories Ø A: Close family members Ø B: Digitally Gated Community Ø C: Access to non-curated spaces (Chain of digital custody certification, Not Blocked By Default) Ø D: Fast Media certification Ø E: Access to Space/Time Browser

Proceedings CoNext 2050 Nuuk, Greenland

Thank You!