The contents of this legacy page are no longer maintained nor supported, and are made available only for historical purposes.

NSFNET: Internet Atlas Gallery

NSFNET growth until 1995

Donna Cox and Robert Patterson
The National Center for Supercomputing Applications (NCSA),
National Science Foundation's Supercomputer Centers Program
University of Illinois at Urbana-Champaign

URL: http://www.ncsa.illinois.edu

NSFNET was a series of networks coordinated by NFS, created to serve research and education purposes. Based on ARPANET protocols, NFSNET served as the backbone of a growing internet up until April 1995. The role of the backbone has since been handed over to several commercial carriers, dissolving the notion of a single national Internet backbone.

Visualization Thumbnail Analysis
Click Image for Full-Size Visualization Back to Gallery

Additional Content

Internet Atlas Gallery

NSFNET growth until 1995

Donna Cox and Robert Patterson
The National Center for Supercomputing Applications (NCSA),
National Science Foundation's Supercomputer Centers Program
University of Illinois at Urbana-Champaign

URL: http://www.ncsa.illinois.edu
Visualization Thumbnail Credits / Contact Info

Authors:

  • Donna Cox
  • Robert Patterson
  • NCSA
  • Data provided by:
    • Merit Network, Inc.
    • NCAR
    • NASA JPL
Contact Info:
For further information, send email to:
media@ncsa.uiuc.edu
Additional Media:
Video depicting two year's of growth of NSFNET (1992-94?) using this visualization:
QuickTime Movie
MPEG Movie
Click Image for Full-Size Visualization Back

Internet Atlas Gallery

NSFNET growth until 1995

Donna Cox and Robert Patterson
The National Center for Supercomputing Applications (NCSA),
National Science Foundation's Supercomputer Centers Program
University of Illinois at Urbana-Champaign

URL: http://www.ncsa.illinois.edu
Visualization Thumbnail Data
Data Sources:
(Source1): Traffic bandwidth in bytes passing through each of the backbone nodes was provided by Merit Network, Inc., NCAR, and NASA JPL. During December 1994, total inbound traffic on the backbone was 18.5 trillion bytes.
Data Aggregation:
(Aggregate1): Networks that exist in the same city and report to the same backbone node are aggregated into a single colored line. During December 1994, (for byte traffic into the ANS/NSFNET T3 backbone from its client networks), 24,435 domestic client networks were aggregated into 12,177 virtual traffic connections.
Data Filters:
(Filter1): Intermediate connections are removed so that only a hypothetical direct connection between each node of the backbone and all of its clients is displayed.
Click Image for Full-Size Visualization Back

Internet Atlas Gallery

NSFNET growth until 1995

Donna Cox and Robert Patterson
The National Center for Supercomputing Applications (NCSA),
National Science Foundation's Supercomputer Centers Program
University of Illinois at Urbana-Champaign

URL: http://www.ncsa.illinois.edu
Visualization Thumbnail Evaluation
Strengths:
(Strength1): Visualization stresses the hierarchy of connections between backbone systems. Backbone nodes are prominently displayed above the geographical map, with all their respective clients below.
(Strength2): High level connectivity is readily perceivable when separated from the clutter of backbone-client connections.
Issues:
(Issue1): Only aggregated clients are displayed, and specific paths from backbone nodes to their clients are not known, therefore distances shown between the backbone and client nodes may be misleading.
(Issue2): Many backbone nodes serve distant clients across the continental US. This logical connectivity blurs attempts to correlate traffic with the geographical organization of the backbone and client nodes. While high connectivity is depicted, the visualization tends to be extremely busy and at times difficult to interpret.
Click Image for Full-Size Visualization Back

Internet Atlas Gallery

NSFNET growth until 1995

Donna Cox and Robert Patterson
The National Center for Supercomputing Applications (NCSA),
National Science Foundation's Supercomputer Centers Program
University of Illinois at Urbana-Champaign

URL: http://www.ncsa.illinois.edu
Visualization Thumbnail About the Visualization
Visualization Techniques:
(Technique1): Backbone nodes are elevated above the geographical map to imply that all data transfer must first travel up to the backbone, then back down to its final destination elsewhere on the map. Backbone nodes are also represented as red spheres at their geographic position on the map of the USA. Client nodes appear as white spheres placed geographically on the map.
(Technique2): Each line between the backbone nodes and their clients indicates geographical location as well as traffic volume.
Key Visualization Mappings:
(Mapping1): Connections are colored by volume of traffic, according to the color key below the image. Purple lines show the least traffic while white lines show the most.
(Mapping2): Node type is depicted by color. Red spheres represent backbone nodes. White spheres represent clients.
Click Image for Full-Size Visualization Back
Published