Stampede at the Texas Advanced Computing Center (TACC) is one of the most powerful high performance computing systems in the world for open science research, and Weather Research and Forecasting Model (WRF) is a parallel mesoscale weather model widely used for research, forecasts, and super computer benchmarking. In this project, TACC, Raytheon, and NCAR conducted a highly localized WRF simulation on Stampede for a nested domain centered at Chicago’s O’Hare international Airport. We have obtained an unprecedentedly high resolution and utilized Stampede to the extreme in many aspects. In this presentation, we will describe several advanced techniques for memory management, IO performance tuning, data processing and visualization, which have been deliberately designed and implemented for this project. Animation of this simulation will also be demonstrated at the end of the presentation.
Si Liu received his PhD in applied mathematics at University of Colorado at Boulder in 2009. His PhD research focused on parallel domain decomposition algorithms for inverse elliptic problems.
During 2009-2013, Si worked as a software engineer in Computational Information Systems Laboratory (CISL) at the National Center for Atmospheric Research (NCAR). He provided consulting services and technical support to scientific computing communities, and collaborated with scientists on models development and optimization. He also participated in establishing the Yellowstone supercomputer in the NCAR-Wyoming Supercomputing Center (NWSC) for climate research.
Si joined the High Performance Computing Group at the Texas Advanced Computing Center (TACC) as a Research Associate in 2013. He is now collaborating with UT research groups, XSEDE community, and many corporations. His current research interests include parallel computing, IO performance, test management, benchmark, and optimization.
| Attachment | Size |
|---|---|
 SEA-2014-SiLiu.pdf | 5.99 MB |
