MAGIC Gigabit Testbed and MAGIC-II

The MAGIC project begain in the early 1990s under the principle that an increasing number of applications utilize powerful computing resources that are distributed over local- and wide-area networks (LANs and WANs). In such an environment, the full potential of these resources can be realized only if the network speeds are sufficient to support the processing demands. WANs at that time operated at megabit-per-second (Mbps) speeds and therefore could limit the overall performance of these distributed systems. Gigabit-per-second WANs promised the next major advance in computing and communications: powerful, geographically distributed computing resources with high-speed access to remote and time-critical data sources. Such networks would allow researchers to develop distributed, interactive applications with massive and real-time data requirements. Furthermore, they would allow data from multiple sources to be integrated for use by these applications, with neither the data sources nor the users colocated with the computing resources. Many challenges needed to be addressed before the benefits of gigabit WANs could be achieved. Someof these include:

The MAGIC (Multidimensional Applications and Gigabit Internetwork Consortium) project was established to develop a very high-speed, wide-area networking testbed that would address these challenges and demonstrate real-time, interactive exchange of data at gigabit-per-second (Gbps) rates among multiple distributed servers and clients.


Participants in the project included organizations from government, industry, and academia. It was one of the six "information superhighway" projects funded by the Information Technology Office (ITO) of the Defense Advanced Research Projects Agency (DARPA) and the National Science Foundation (NSF). MAGIC participants included:

The MAGIC testbed consisted of the following three components:

  1. An interactive, real-time, terrain visualization application
  2. A distributed image server system with performance sufficient to support the terrainvisualization application
  3. A high-speed internetwork to link the computing resources required for real-time rendering of the terrain

The terrain visualization application, known as TerraVision, was designed to allow a user to view andnavigate through a representation of a landscape. The initial landscape was created from aerial images of the U.S. Army National Training Center (NTC) in Fort Irwin, California. As exercises were conducted, the locations of vehicles were superimposed on the view of the terrain and updated in real time. TerraVision is of direct interest to the U.S. Army since the ability of acommander to see the battlefield, and to share a common view of the battlefield with his command, is critical to effective command and control. TerraVision requires very large amounts of data in real time, transferred at both very bursty and high steady rates, and has network throughput as its major limiting factor. Steady traffic occurs when a user moves smoothly through the terrain, whereas bursty traffic occurs when the user jumps ("teleports") to a new position.

The following documentation is available on the MAGIC project.