OIL & GAS
Writer
Storm-tracking System Relies on Grid Computing
By Ann Bednarz, Network World Fusion - A $40 million research center launched Wednesday is designed to change the way scientists detect and predict weather hazards such as tornadoes and flash floods. The Engineering Research Center for Collaborative Adaptive Sensing of the Atmosphere (CASA), based at the University of Massachusetts in Amherst, is working to create a network of radar dishes that will collect atmospheric data to enable earlier and more accurate forecasts of weather emergencies, participants say. The radars will be linked by an IBM-based grid computing infrastructure that pools the resources of geographically dispersed systems. 
CASA hopes to overcome a shortcoming of existing forecasting and warning systems, which have difficulty monitoring conditions in the lower atmosphere. Today’s systems use data from high-power, long-range radars that have limited ability to observe the lower part of the atmosphere because of earth’s curvature, CASA says. It plans to overcome the curvature issue and obstructions such as mountains and buildings by setting up low-cost, dense networks of small radars that operate at short range. As the weather changes, the radars will automatically adjust their sensing strategies in response, CASA says. IBM is providing hardware and software to support the research center’s efforts. The grid system will include IBM blade servers running Linux. IBM also is supplying WebSphere Application Server software, DB2 database software and Rational application development tools. Team members will use the IBM gear to help integrate various systems as users from academic, public and private sectors access multiple applications on the network. A key issue for CASA is deciding how to automate the allocation of computing and storage resources on the fly. It will be weather dependent – computing power must be allocated to areas where demand is greatest due to atmospheric conditions, participants say. The first field test of CASA will take place in mid-2005 in Oklahoma and will cover about 20% of the state in an area that experiences about 22 tornadoes each year. The second test will be in Houston, where CASA will deploy a system to predict floods more accurately. A third test is planned in Puerto Rico to improve monitoring of hurricanes as they approach land. UMass Amherst is leading the CASA team, which includes engineers, meteorologists, atmospheric scientists and sociologists from partner institutions. Educational partners include University of Oklahoma, Colorado State University, and University of Puerto Rico. In addition to IBM, CASA’s industry partners include Raytheon, M/A-COM, Vaisala, Vieux and Associates, Telephonics, and The Weather Channel. Government partners include the National Oceanic and Atmospheric Administration (NOAA) National Severe Storms Laboratory, the National Weather Service and Oak Ridge National Laboratory. UMass and its partners will operate CASA with the help of $40 million in funding over five years. The funding includes a $17 million grant from the National Science Foundation and $5 million from the Commonwealth of Massachusetts.
CASA hopes to overcome a shortcoming of existing forecasting and warning systems, which have difficulty monitoring conditions in the lower atmosphere. Today’s systems use data from high-power, long-range radars that have limited ability to observe the lower part of the atmosphere because of earth’s curvature, CASA says. It plans to overcome the curvature issue and obstructions such as mountains and buildings by setting up low-cost, dense networks of small radars that operate at short range. As the weather changes, the radars will automatically adjust their sensing strategies in response, CASA says. IBM is providing hardware and software to support the research center’s efforts. The grid system will include IBM blade servers running Linux. IBM also is supplying WebSphere Application Server software, DB2 database software and Rational application development tools. Team members will use the IBM gear to help integrate various systems as users from academic, public and private sectors access multiple applications on the network. A key issue for CASA is deciding how to automate the allocation of computing and storage resources on the fly. It will be weather dependent – computing power must be allocated to areas where demand is greatest due to atmospheric conditions, participants say. The first field test of CASA will take place in mid-2005 in Oklahoma and will cover about 20% of the state in an area that experiences about 22 tornadoes each year. The second test will be in Houston, where CASA will deploy a system to predict floods more accurately. A third test is planned in Puerto Rico to improve monitoring of hurricanes as they approach land. UMass Amherst is leading the CASA team, which includes engineers, meteorologists, atmospheric scientists and sociologists from partner institutions. Educational partners include University of Oklahoma, Colorado State University, and University of Puerto Rico. In addition to IBM, CASA’s industry partners include Raytheon, M/A-COM, Vaisala, Vieux and Associates, Telephonics, and The Weather Channel. Government partners include the National Oceanic and Atmospheric Administration (NOAA) National Severe Storms Laboratory, the National Weather Service and Oak Ridge National Laboratory. UMass and its partners will operate CASA with the help of $40 million in funding over five years. The funding includes a $17 million grant from the National Science Foundation and $5 million from the Commonwealth of Massachusetts.