swPHoToNs: Toward trillion‐body‐scale cosmological N‐body simulations on Sunway TaihuLight supercomputer
Abstract Cosmological N‐body simulations have been essential for astronomers to study the formation of nonlinear structures and hypotheses of dark matter, dark energy, etc. The scale of the problem naturally leads to extreme scenarios with billions or even trillions of particles, thus demanding mass...
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2025-03-01
|
| Series: | Engineering Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1002/eng2.12640 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849340016679976960 |
|---|---|
| author | Zhao Liu Qiao Wang Xiaojing Lv Xu Liu Hongsong Meng Guanghui Zhu Liang Gao Haohuan Fu Guangwen Yang |
| author_facet | Zhao Liu Qiao Wang Xiaojing Lv Xu Liu Hongsong Meng Guanghui Zhu Liang Gao Haohuan Fu Guangwen Yang |
| author_sort | Zhao Liu |
| collection | DOAJ |
| description | Abstract Cosmological N‐body simulations have been essential for astronomers to study the formation of nonlinear structures and hypotheses of dark matter, dark energy, etc. The scale of the problem naturally leads to extreme scenarios with billions or even trillions of particles, thus demanding massive computational power and highly efficient algorithms. In this paper, we present swPHoToNs, a Particle‐Mesh (PM) and Fast Multipole Method‐ (FMM) based code that can perform cosmological simulations with trillions of particles efficiently on the Sunway TaihuLight supercomputer. Our design includes three novel optimizations: (1) a multilevel domain decomposition and dynamic load‐balancing scheme; (2) a pipeline strategy for tree traversal and gravity calculation; (3) optimizations for both computation and MPI kernels with consideration of the hardware features. We manage to conduct cosmological simulations which contain up to 1.6 trillion particles, obtaining a sustained performance of 56.3 PFlops with a weak‐scaling parallel efficiency of 80.9% and a computational efficiency of 44.9%. |
| format | Article |
| id | doaj-art-b06e8f2da0774b85800dbe38fe36c406 |
| institution | Kabale University |
| issn | 2577-8196 |
| language | English |
| publishDate | 2025-03-01 |
| publisher | Wiley |
| record_format | Article |
| series | Engineering Reports |
| spelling | doaj-art-b06e8f2da0774b85800dbe38fe36c4062025-08-20T03:44:00ZengWileyEngineering Reports2577-81962025-03-0173n/an/a10.1002/eng2.12640swPHoToNs: Toward trillion‐body‐scale cosmological N‐body simulations on Sunway TaihuLight supercomputerZhao Liu0Qiao Wang1Xiaojing Lv2Xu Liu3Hongsong Meng4Guanghui Zhu5Liang Gao6Haohuan Fu7Guangwen Yang8Department of Computer Science and Technology Tsinghua University Beijing ChinaKey Laboratory of Computational Astrophysics, National Astronomical Observatories Chinese Academy of Sciences Beijing ChinaDepartment of Parallel Optimization National Supercomputing Center in Wuxi Wuxi ChinaDepartment of Parallel Optimization National Supercomputing Center in Wuxi Wuxi ChinaDepartment of Parallel Optimization National Supercomputing Center in Wuxi Wuxi ChinaDepartment of Government and Enterprise Business Huirong Electronic Systems Engineering Co., Ltd. Shanghai ChinaKey Laboratory of Computational Astrophysics, National Astronomical Observatories Chinese Academy of Sciences Beijing ChinaDepartment of Earth System Science Tsinghua University Beijing ChinaDepartment of Computer Science and Technology Tsinghua University Beijing ChinaAbstract Cosmological N‐body simulations have been essential for astronomers to study the formation of nonlinear structures and hypotheses of dark matter, dark energy, etc. The scale of the problem naturally leads to extreme scenarios with billions or even trillions of particles, thus demanding massive computational power and highly efficient algorithms. In this paper, we present swPHoToNs, a Particle‐Mesh (PM) and Fast Multipole Method‐ (FMM) based code that can perform cosmological simulations with trillions of particles efficiently on the Sunway TaihuLight supercomputer. Our design includes three novel optimizations: (1) a multilevel domain decomposition and dynamic load‐balancing scheme; (2) a pipeline strategy for tree traversal and gravity calculation; (3) optimizations for both computation and MPI kernels with consideration of the hardware features. We manage to conduct cosmological simulations which contain up to 1.6 trillion particles, obtaining a sustained performance of 56.3 PFlops with a weak‐scaling parallel efficiency of 80.9% and a computational efficiency of 44.9%.https://doi.org/10.1002/eng2.12640cosmological N‐body simulationFMMparallel scalabilitySunway TaihuLight |
| spellingShingle | Zhao Liu Qiao Wang Xiaojing Lv Xu Liu Hongsong Meng Guanghui Zhu Liang Gao Haohuan Fu Guangwen Yang swPHoToNs: Toward trillion‐body‐scale cosmological N‐body simulations on Sunway TaihuLight supercomputer Engineering Reports cosmological N‐body simulation FMM parallel scalability Sunway TaihuLight |
| title | swPHoToNs: Toward trillion‐body‐scale cosmological N‐body simulations on Sunway TaihuLight supercomputer |
| title_full | swPHoToNs: Toward trillion‐body‐scale cosmological N‐body simulations on Sunway TaihuLight supercomputer |
| title_fullStr | swPHoToNs: Toward trillion‐body‐scale cosmological N‐body simulations on Sunway TaihuLight supercomputer |
| title_full_unstemmed | swPHoToNs: Toward trillion‐body‐scale cosmological N‐body simulations on Sunway TaihuLight supercomputer |
| title_short | swPHoToNs: Toward trillion‐body‐scale cosmological N‐body simulations on Sunway TaihuLight supercomputer |
| title_sort | swphotons toward trillion body scale cosmological n body simulations on sunway taihulight supercomputer |
| topic | cosmological N‐body simulation FMM parallel scalability Sunway TaihuLight |
| url | https://doi.org/10.1002/eng2.12640 |
| work_keys_str_mv | AT zhaoliu swphotonstowardtrillionbodyscalecosmologicalnbodysimulationsonsunwaytaihulightsupercomputer AT qiaowang swphotonstowardtrillionbodyscalecosmologicalnbodysimulationsonsunwaytaihulightsupercomputer AT xiaojinglv swphotonstowardtrillionbodyscalecosmologicalnbodysimulationsonsunwaytaihulightsupercomputer AT xuliu swphotonstowardtrillionbodyscalecosmologicalnbodysimulationsonsunwaytaihulightsupercomputer AT hongsongmeng swphotonstowardtrillionbodyscalecosmologicalnbodysimulationsonsunwaytaihulightsupercomputer AT guanghuizhu swphotonstowardtrillionbodyscalecosmologicalnbodysimulationsonsunwaytaihulightsupercomputer AT lianggao swphotonstowardtrillionbodyscalecosmologicalnbodysimulationsonsunwaytaihulightsupercomputer AT haohuanfu swphotonstowardtrillionbodyscalecosmologicalnbodysimulationsonsunwaytaihulightsupercomputer AT guangwenyang swphotonstowardtrillionbodyscalecosmologicalnbodysimulationsonsunwaytaihulightsupercomputer |