RSS|
CENTERFOLDProcessing And Research At Marine SchoolTo access a gif file of the Centerfold graphic, click here. It may take some time to load in your browser! by Linda Nicastro
T
he University of Miami's Rosentiel School of Marine and Atmospheric Science (RSMAS) in Miami, Fla., uses high-speed networking to process satellite data on global sea surface temperatures and other ocean phenomena.
Daily, the school relies on an FDDI network to move 100 GB of input data and 20 GB of output using a rules-based processing system that controls data, resource allocation and tasking. All told, RSMAS uses three operational networks--including Ethernet segments, FDDI and ATM--to support its research efforts.
The growth of the RSMAS network has paralleled an increase in the group's large-scale processing needs. Today, FDDI is used as the primary network, but RSMAS is migrating processing machines to ATM to accommodate an increasing data flow and facilitate collaborative research. The school's research vision is interdisciplinary and based on cooperative interaction among scientists at many institutions. RSMAS is also focused on global change that entails studying the Earth as a system, not as isolated pieces. With networking and ATM, RSMAS sees improved opportunities for sharing computing resources, exchanging data, and using applications such as voice and video to enhance research and make massive amounts of data more accessible.
While multiple networks present a unique set of challenges, such as efficient troubleshooting, in the end it is the continuous flow of data that matters most. "One of the greatest challenges in our networking is assuring that the data flow is continuously available," said Peter Evans, senior research associate at RSMAS.
|
