TY - GEN
T1 - Toward Exascale Computation for Turbomachinery Flows
AU - Fu, Yuhang
AU - Shen, Weiqi
AU - Cui, Jiahuan
AU - Zheng, Yao
AU - Yang, Guangwen
AU - Liu, Zhao
AU - Zhang, Jifa
AU - Ji, Tingwei
AU - Xie, Fangfang
AU - Lv, Xiaojing
AU - Liu, Hanyue
AU - Liu, Xu
AU - Liu, Xiyang
AU - Song, Xiaoyu
AU - Tao, Guocheng
AU - Yan, Yan
AU - Tucker, Paul
AU - Miller, Steven
AU - Luo, Shirui
AU - Koric, Seid
AU - Zheng, Weimin
N1 - We acknowledge the support from the National Nature Science Foundation of China with grant number 52106060 and 92152202. The opensourced HiFiLES and SU2 project at Stanford University and the PyFR project at Imperial College London inspired some of our algorithm development in this work, which are also acknowledged here. We also thank Will Trojak at IBM for useful technical discussions.
PY - 2023/11/12
Y1 - 2023/11/12
N2 - A state-of-the-art large eddy simulation code has been developed to solve compressible flows in turbomachinery. The code has been engineered with a high degree of scalability, enabling it to effectively leverage the many-core architecture of the new Sunway system. A consistent performance of 115.8 DP-PFLOPs has been achieved on a high-pressure turbine cascade consisting of over 1.69 billion mesh elements and 865 billion Degree of Freedoms (DOFs). By leveraging a high-order unstructured solver and its portability to large heterogeneous parallel systems, we have progressed towards solving the grand challenge problem outlined by NASA [1], which involves a time-dependent simulation of a complete engine, incorporating all the aerodynamic and heat transfer components.
AB - A state-of-the-art large eddy simulation code has been developed to solve compressible flows in turbomachinery. The code has been engineered with a high degree of scalability, enabling it to effectively leverage the many-core architecture of the new Sunway system. A consistent performance of 115.8 DP-PFLOPs has been achieved on a high-pressure turbine cascade consisting of over 1.69 billion mesh elements and 865 billion Degree of Freedoms (DOFs). By leveraging a high-order unstructured solver and its portability to large heterogeneous parallel systems, we have progressed towards solving the grand challenge problem outlined by NASA [1], which involves a time-dependent simulation of a complete engine, incorporating all the aerodynamic and heat transfer components.
KW - Exascale computing
KW - flux reconstruction method
KW - heterogeneous many-core system
KW - large eddy simulation
KW - Sunway supercomputer
KW - turbomachinery
UR - http://www.scopus.com/inward/record.url?scp=85190420089&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85190420089&partnerID=8YFLogxK
U2 - 10.1145/3581784.3627040
DO - 10.1145/3581784.3627040
M3 - Conference contribution
AN - SCOPUS:85190420089
T3 - International Conference for High Performance Computing, Networking, Storage and Analysis, SC
BT - SC '23: Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis
PB - Association for Computing Machinery
T2 - 2023 International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2023
Y2 - 12 November 2023 through 17 November 2023
ER -