Four UW Research Projects Chosen for Third Cycle of Supercomputer Use
August 23, 2013 — Four computational research projects led by University of Wyoming faculty members aimed at providing insights on key issues in Wyoming and the West have been selected to use the NCAR-Wyoming Supercomputing Center (NWSC) in Cheyenne over the next year.
The research projects were chosen and approved by a panel appointed by the Computer Information Science Laboratory at the National Center for Atmospheric Research (NCAR). The computational science research projects will use approximately 10.2 million core processor hours on the Cheyenne supercomputer.
“These allocations will support major research projects at UW that will enable scientists to better understand the behavior of atmospheric, biological, geological and hydrological systems that are related to important societal and, indeed, Wyoming-specific issues, such as: How can Wyoming’s water be most effectively managed to meet municipal, agricultural, ecological, energy related and recreational needs? How do aerosols from forest fires, volcanic eruptions and air pollution affect cloud formation and precipitation? What is the most effective design for a wind turbine?” says Bryan Shader, UW’s special assistant to the vice president of research and economic development, and a mathematics professor.
Shader also serves as co-chair of the Wyoming-NCAR Alliance Resource Allocations Panel (WRAP), a group that chooses which projects receive core-hour allocations on Yellowstone, the nickname for the Cheyenne supercomputer.
NWSC Veterans Plan Water Research
Two of the projects -- both focusing on water -- will be led by UW faculty members with previous experience using the supercomputer. They are:
--Craig Douglas, a UW School of Energy Resources (SER) professor of mathematics and director for the Institute for Scientific Computations, “CI-Water Petascale Computation of Model for the Upper Colorado River Basin.” A comprehensive model of the upper Colorado River Basin -- at a resolution 100 times higher than currently available -- will be created. This model will better enable policy and management decisions regarding water in the basin.
This continues a project Douglas started last year with Fred Ogden, the Cline Distinguished Chair in UW’s Department of Civil & Architectural Engineering, and Haub School of Environmental and Natural Resources.
Douglas's work is part of UW's research efforts with the CI-WATER project, which is funded with an EPSCoR Research Infrastructure Improvement Program (RII) Track-2 cooperative agreement that focuses on the hydrology of the Colorado River Basin.
The CI-WATER project is a joint collaboration among UW, the University of Utah, Utah State University and Brigham Young University. Cooperators include the U.S. Army Corps of Engineers and NCAR. EPSCoR Track-2 cooperative agreements, distributed through the NSF, are designed to provide research funding to states, including Wyoming, that typically receive lesser amounts of NSF research and development funding.
Ogden and Norman Jones, a professor in the Civil & Environmental Engineering Department at Brigham Young University, are the project leaders.
The project is funded by a National Science Foundation (NSF) grant under the award title, “Collaborative Research: CI-WATER, Cyberinfrastructure to Advance High Performance Water Resource Modeling.”
--Po Chen, a UW SER associate professor of geology and geophysics, “Geophysical and Hydrological Integrated Modeling for Wyoming Water Resource Management.” This research project is tied to the Wyoming EPSCoR $20 million Track-I water grant.
“Reliable surface-subsurface hydrologic modeling and forecasting require detailed knowledge of subsurface porous media,” Chen says. “An important research objective of the Center for Near-Surface Imaging and Science (CeNSIS) is to develop the capability that combines different types of geophysical observations to image hydrologically relevant properties of the subsurface at a meter-scale resolution in 3-D and integrates such images into operational hydrological modeling and forecasting.”
Anne Sylvester, a UW professor of molecular biology, serves as the principal investigator. Steve Holbrook, a UW professor of geology and geophysics, and Scott Miller, a UW associate professor of ecosystem science & management, serve as co-principal investigators on the Track-I water grant.
This project has been funded by the NSF Office of International and Integrative Activities under the title, “Water in a Changing West: The Wyoming Center for Environmental Hydrology and Geophysics.”
Chen previously used the supercomputer to model seismic events to improve warning systems for earthquakes.
New Faces, Familiar Subjects
--Xiaohong Liu, a UW associate professor of atmospheric science and the Wyoming Excellence Chair in Climate Science, “Uncertainty Quantification of Aerosol Indirect Effects to Parameters in Cloud Macrophysics, Convection and Turbulence Parameterizations in NCAR CAM5/CESMI.”
A new UW faculty member, Liu will study how aerosols -- created from soil, dust, volcanic ash and the burning of fossil fuels -- affect the atmosphere, and develop prediction models for the future.
“The presence of aerosols affects cloud formation which, in turn, can impact climate,” Shader says. “Professor Liu and his collaborators are trying to understand that. They want to have more accurate models to model regional climate 10 years out.”
This project is funded by NSF and the Department of Education under the title “Quantifying the Uncertainties of Aerosol Indirect Effects and Impacts on Decadal-Scale Climate Variability in NCAR CAM5 and CESM1.”
--Jay Sitaraman, a UW assistant professor of mechanical engineering, “Computational Fluid Dynamic Simulations Using a Hybrid CPU/GPU Paradigm.”
“Professor Sitaraman and his team are developing an entire suite of computational tools that will allow people to make engineering decisions on how to build (wind) turbines or helicopter rotors,” Shader says.
Flow fields on wind turbines rank among the most challenging, which makes them computationally demanding to simulate, according to a portion of the project overview. Accurate representation of the flow field around a turbine or rotor is essential in assessing overall performance, especially for new designs.
Using a mix of graphic processing units (GPUs), such as those used in video games, and central processing units (CPUs), Sitaraman will be able to speed up his calculations ten-fold, Shader says.
“Any computations that can be performed faster, and with the same accuracy, allows one to simulate over a broader range of parameters or with higher resolution, or incorporate more physics into the model to increase utility of the model,” Shader says.
Dominic Chandar, a UW post-doctoral research associate in mechanical engineering, and Dimitri Mavriplis, a UW professor of mechanical engineering, will serve as co-investigators on the project.
This project is funded by the Army Research Office, under the title, “Advanced Overset Grid Methods for Massively Parallel Rotary Wing Computations.”
For these four projects, UW researchers have had access to Yellowstone since Aug. 1, Shader says.
“They will have a year to work on their research projects, with a possibility of renewal allocation,” he adds.
NWSC is the result of a partnership among the University Corporation for Atmospheric Research (UCAR), the operating entity for NCAR; the University of Wyoming; the state of Wyoming; Cheyenne LEADS; the Wyoming Business Council; and Cheyenne Light, Fuel & Power. The NWSC is operated by NCAR under sponsorship of the NSF.
The NWSC contains one of the world's most powerful supercomputers (1.5 petaflops, which is equal to 1.5 quadrillion mathematical operations per second) dedicated to improving scientific understanding of climate change, severe weather, air quality and other vital atmospheric science and geoscience topics. The center also houses a premier data storage (11 petabytes) and archival facility that holds historical climate records and other information.
Craig Douglas, a UW SER professor of mathematics, will use the supercomputer to create a comprehensive model of the upper Colorado River Basin. This model will better enable policy and management decisions regarding water in the basin.