Infomax: An Auto-summarizing Information Delivery Protocol Tarek - - PowerPoint PPT Presentation

▶
infomax an auto summarizing information delivery protocol
SMART_READER_LITE
LIVE PREVIEW

Infomax: An Auto-summarizing Information Delivery Protocol Tarek - - PowerPoint PPT Presentation

Jan 23, 2024 34 likes •274 views

Cyber-physical Computing Group Infomax: An Auto-summarizing Information Delivery Protocol Tarek Abdelzaher University of Illinois at Urbana Champaign Cyber-physical Computing Group The Applicat cation n Land ndscape What trends

slide-1
SLIDE 1

Infomax: An Auto-summarizing Information Delivery Protocol

Tarek Abdelzaher

University of Illinois at Urbana Champaign

Cyber-physical Computing Group

slide-2
SLIDE 2

The Applicat cation n Land ndscape

What trends characterize big future shifts in the application landscape?

Cyber-physical Computing Group

slide-3
SLIDE 3

The Age of Data

 Ubiquitous digital sources:

Cell-phones Sportsware Glucose monitor Smart Meter GPS Pulse

  • ximeter

Kinect Smart Eyeware Smart Watch Smart Sportsware Wii

slide-4
SLIDE 4

The Age of (Democratized) Broadcast

 Ubiquitous digital sources:  Unprecedented dissemination opportunities

Cell-phones Sportsware Glucose monitor Smart Meter GPS Pulse

  • ximeter

Kinect Smart Eyeware Smart Watch Smart Sportsware Wii

slide-5
SLIDE 5

 Broadcast in the 20th century:

The Age of (Democratized) Broadcast

slide-6
SLIDE 6

 Broadcast in the 20th century:  Broadcast today:

The Age of (Democratized) Broadcast

slide-7
SLIDE 7

Implication

The rate of data production will increasingly outpace application data consumption needs

slide-8
SLIDE 8

A Paradigm Shift Information Retrieval  Information Distillation

Matching Data at Sources Data Received at Sinks

TCP (Transport Control) 1 bit sent  1 bit received

Current Applications (HTTP , FTP , VoIP , etc.)

slide-9
SLIDE 9

A Paradigm Shift Information Retrieval  Information Distillation

Matching Data at Sources Data Received at Sinks

TCP (Transport Control) 1 bit sent  1 bit received

Current Applications (HTTP , FTP , VoIP , etc.)

Matching Data at Sources

Data Received at Sinks

Reduction, subsampling, summarization, …

Future Applications

slide-10
SLIDE 10

A Paradigm Shift Information Retrieval  Information Distillation

Matching Data at Sources

Data Received at Sinks

Reduction, subsampling, summarization, …

Future Applications

In NDN, the network is aware of (i) application-level object boundaries and (ii) topological relations between

  • bject names.

This knowledge can be exploited for data volume reduction

slide-11
SLIDE 11

A “Tra Trans nsport rt Laye yer” r” Solut ution

Infomax

Cyber-physical Computing Group

slide-12
SLIDE 12

Infomax:

A Novel Data Retrieval API

 Get (/subtree)  Semantics:

 Retrieve a representative sampling of data

  • bjects under /subtree
slide-13
SLIDE 13

Infomax:

A Novel Data Retrieval API

 Get (/subtree)  Semantics:

 Retrieve a representative sampling of data

  • bjects under /subtree

 Note 1: Representative sampling

 minimally redundant

slide-14
SLIDE 14

Infomax:

A Novel Data Retrieval API

 Get (/subtree)  Semantics:

 Retrieve a representative sampling of data

  • bjects under /subtree

 Note 1: Representative sampling

 minimally redundant

slide-15
SLIDE 15

Infomax:

A Novel Data Retrieval API

 Get (/subtree)  Semantics:

 Retrieve a representative sampling of data

  • bjects under /subtree

 Note 1: Representative sampling

 minimally redundant

 Note 2: Longer shared prefix between objects

 more semantic redundancy

slide-16
SLIDE 16

A Content Transmission Prioritization Policy

Least shared prefix first

slide-17
SLIDE 17

A Content Transmission Prioritization Policy

Least shared prefix first

Tie? Take leftmost branch

slide-18
SLIDE 18

A Content Transmission Prioritization Policy

Least shared prefix first

Note: Reduces approximately a breadth-first traversal of the content under /subtree

/subtree Tie? Take leftmost branch

slide-19
SLIDE 19

A Content Transmission Prioritization Policy

Least shared prefix first

Note: Reduces approximately a breadth-first traversal of the content under /subtree

1 /subtree Tie? Take leftmost branch

slide-20
SLIDE 20

A Content Transmission Prioritization Policy

Least shared prefix first

Note: Reduces approximately a breadth-first traversal of the content under /subtree

1 2 /subtree Tie? Take leftmost branch

slide-21
SLIDE 21

A Content Transmission Prioritization Policy

Least shared prefix first

Note: Reduces approximately a breadth-first traversal of the content under /subtree

1 3 2 /subtree Tie? Take leftmost branch

slide-22
SLIDE 22

A Content Transmission Prioritization Policy

Least shared prefix first

Note: Reduces approximately a breadth-first traversal of the content under /subtree

1 3 2 4 /subtree Tie? Take leftmost branch

slide-23
SLIDE 23

Protocol Design

Consumer Producer

(advertises / subt

bt ree)

Get (/ subt ree)

Requests all content under subtree

Repeat until the consumer gets enough data

Request objects on list in

  • rder (fewer requests =

more summarization)

slide-24
SLIDE 24

Evaluation

 Currently, evaluation is underway on

testbed.

1 2 3 4 5 6 7 8 1 11 21 31 41 51 61 71 81 91

The length of shared prefix the number of nodes in a name tree

TCP InfoMax

0% 20% 40% 60% 80% 100% 1 6 11 16 21

the number of requests

Level1 Level2 Level3