Research by Tsuyoshi Ichimura, Kohei Fujita, Takuma Yamaguchi, Muneo Hori, Lalith Maddegedara was selected as Gordon Bell Prize Finalist.
◆This has been press released on 16th November, 2018 :https://www.u-tokyo.ac.jp/focus/en/articles/z0508_00018.html
Paper: A Fast Scalable Implicit Solver for Nonlinear Time-Evolution Earthquake City Problem on Low-Ordered Unstructured Finite Elements with Artificial Intelligence and Transprecision Computing
Authors：Tsuyoshi Ichimura1, 2, 3, Kohei Fujita1, 3, Takuma Yamaguchi1, Akira Naruse4, Jack C. Wells5, Thomas C. Schulthess6, Tjerk P. Straatsma5, Christopher J. Zimmer5, Maxime Martinasso6, Kengo Nakajima7,3, Muneo Hori1,3,Lalith Maddegedara1,3.
（1: Earthquake Research Institute · Graduate School of Engineering, Department of Social Infrastructure Studies, Univ.Tokyo 2: RIKEN Innovation Intelligence Integration Research Center, 3: RIKEN Computational Science Research Center, 4: NVIDIA Corporation, 5: Oak Ridge National Laboratory, 6: Swiss National Supercomputing Centre, 7: Information technology Center, Univ.Tokyo）
The Gordon Bell Prize is awarded each year from ACM (Association for Computing Machinery) to recognize outstanding achievement in high-performance computing. It is known as one of the most prestigious international prize in the field of high-performance computing. The prize winner will be announced at SC18(International Conference for High Performance Computing, Networking, Storage, and Analysis) starting on 11th November in Dallas, Texas.
The research is on Ultra High Resolution Urban Earthquake simulator which simulated an earthquake wave—accelerated using artificial intelligence (AI) and a computational technique called transprecision computing.
Also the shaking of the ground is coupled with underground and above-ground building structures in high resolution analysis for the first time, which is expected to improve reliability of earthquake simulations.
（Oak Ridge National Laboratory）