@article{Yan2024, 
author = {Yu-Jin Yan and Hai-Bo Li and Tong Zhao and Lin-Wang Wang and Lin Shi and Tao Liu and Guang-Ming Tan and Wei-Le Jia and Ning-Hui Sun},
title = {10-Million Atoms Simulation of First-Principle Package LS3DF},
year = {2024},
journal = {Journal of Computer Science and Technology},
volume = {39},
number = {1},
pages = {45-62},
keywords = {high-performance computing, electronic structure, deep computing unit, linearly scaling three-dimensional fragment (LS3DF), Sugon supercomputer},
url = {https://www.sciopen.com/article/10.1007/s11390-023-3011-6},
doi = {10.1007/s11390-023-3011-6},
abstract = {The growing demand for semiconductor devices simulation poses a big challenge for large-scale electronic structure calculations. Among various methods, the linearly scaling three-dimensional fragment (LS3DF) method exhibits excellent scalability in large-scale simulations. Based on algorithmic and system-level optimizations, we propose a highly scalable and highly efficient implementation of LS3DF on the Sugon supercomputer, a domestic supercomputer equipped with deep computing units. In terms of algorithmic optimizations, the original all-band conjugate gradient algorithm is refined to achieve faster convergence, and mixed precision computing is adopted to increase overall efficiency. In terms of system-level optimizations, the original two-layer parallel structure is replaced by a coarse-grained parallel method. Optimization strategies such as multi-stream, kernel fusion, and redundant computation removal are proposed to increase further utilization of the computational power provided by the heterogeneous machines. As a result, our optimized LS3DF can scale to a 10-million silicon atoms system, attaining a peak performance of 34.8 PFLOPS (21.2% of the peak). All the improvements can be adapted to the next-generation supercomputers for larger simulations.}
}