The Edge of Tomorrow
π€ Dark Forests, Relativistic Computing & How to Power a New Internet π
https://fission.codes
Slide 2
The Edge of Tomorrow
Slide 3
The Edge of Tomorrow
Where is the line between smart contract virtual machines and other decentralized computation and data storage systems? What can and should run and be stored on-chain in the future? How do we choose?
Slide 4
The Edge of Tomorrow
Where is the line between smart contract virtual machines and other decentralized computation and data storage systems? What can and should run and be stored on-chain in the future? How do we choose?
The Edge of Tomorrow
Baseline Trajectory AWS 32%
Azure 17% Google 9%
Other 35%
Alibaba 4% IBM 4%
Slide 9
The Edge of Tomorrow
Baseline Trajectory AWS 32%
Azure 17% Google 9%
Other 35%
Alibaba 4% IBM 4%
Slide 10
The Edge of Tomorrow
Slide 11
The Edge of Tomorrow
Nothing less than connecting all of the worldβs users & services. The βHTTPβ storage and compute equivalent: open, interoperable, & everywhere. Must be substantially better than Web 2.0
Slide 12
Consistency & Consensus βOn a Need To Know Basisβ
Slide 13
Consistency & Consensus
Consistency Tradeoffs Global distributed consensus is expensive Time (latency) is a hard physical limit Extremely useful for many things! Mutable pointers, public commitments, public data, broadcast/distribution β¦but not everything
Slide 14
Consistency & Consensus
Very Blurry Pipes
Slide 15
Consistency & Consensus
Very Blurry Pipes Commons
Cloud & Edge
Far Edge
Slide 16
Consistency & Consensus
Very Blurry Pipes Commons
Cloud & Edge
Far Edge
Slide 17
Consistency & Consensus
Very Blurry Pipes Commons
Cloud & Edge
Far Edge
Consistency & Consensus
Growing Toolbox Web3
Serverless
Cloud
Local-First O
ffl
P2P
ine
Slide 23
Consistency & Consensus
Growing Toolbox Web3
Serverless Networked Data Cloud
Commons Networks Local-First Blockchain
O
ffl
P2P
ine
Slide 24
Beyond Light Speed Contending with Raw Physics
Slide 25
Beyond Light Speed
Edge Constraints
Slide 26
Beyond Light Speed
Edge Constraints
Source: Ericsson http://cscn2017.ieee-cscn.org/files/2017/08/Janne_Peisa_Ericsson_CSCN2017.pdf
Slide 27
Beyond Light Speed
Edge Constraints
Source: Ericsson http://cscn2017.ieee-cscn.org/files/2017/08/Janne_Peisa_Ericsson_CSCN2017.pdf
Slide 28
Beyond Light Speed
What 8ms Looks Like
Slide 29
Beyond Light Speed
What 8ms Looks Like Austin β‘ San Francisco Ideal Vacuum π«
Slide 30
Beyond Light Speed
What 8ms Looks Like Austin β‘ San Francisco Ideal Vacuum π«
Austin π (almost) Atlanta Ideal Vacuum π«
Slide 31
Beyond Light Speed
What 8ms Looks Like Austin β‘ San Francisco Ideal Vacuum π«
Austin π (almost) Atlanta Ideal Vacuum π«
Austin π New Orleans Ideal Fibre π§Ά
Slide 32
Beyond Light Speed
What 8ms Looks Like Austin β‘ San Francisco Ideal Vacuum π«
Austin π (almost) Atlanta Ideal Vacuum π«
Austin π New Orleans Ideal Fibre π§Ά
The Dark Forest
Permissionless Auth for Users, Apps, and Machines
Slide 51
The Dark Forest
Slide 52
The Dark Forest
Cryptography is a tool for turning lots of different problems into key management problems Dr. Lea Kissner, Googleβs Global Lead of Privacy Technologies
Slide 53
The Dark Forest
Making Private⦠Public! Binary CBOR
Encrypted Node π AES256
Encrypted Node π
Virtual Node
π Index
Metadata
Encrypted Node π π
π
π
Decentralized Compute
Declarative Invokation
Description of jobs & results Index and/or names for later lookup Streams of results per machine (IPVM & IPLI)
Slide 75
Decentralized Compute
Declarative Invokation
Description of jobs & results
Input Graph
Index and/or names for later lookup Streams of results per machine (IPVM & IPLI)
f
Arguments
Scheduling Con ig, etc
Slide 76
Decentralized Compute
Declarative Invokation
Description of jobs & results
Output Graph
Input Graph
Index and/or names for later lookup Streams of results per machine (IPVM & IPLI)
f
Arguments
Scheduling Con ig, etc
Slide 77
Decentralized Compute
Declarative Invokation
Description of jobs & results
Output Graph
Input Graph
Index and/or names for later lookup
Results
Streams of results per machine (IPVM & IPLI)
f
ff
Arguments
Scheduling Con ig, etc
Managed E ects