Accelerating hartree-fock self-consistent field calculation on C86/DCU heterogenous computing platform

被引：0

作者：

Qi, Ji ^{[1
]}

Zhang, Huimin ^{[1
,2
]}

Shan, Dezun ^{[1
,2
]}

Yang, Minghui ^{[1
,3
]}

机构：

[1] Chinese Acad Sci, Wuhan Inst Phys & Math, Innovat Acad Precis Measurement Sci & Technol, State Key Lab Magnet Resonance Spect & Imaging, Beijing 430071, Peoples R China

[2] Univ Chinese Acad Sci, Beijing 100049, Peoples R China

[3] Huazhong Univ Sci & Technol, Wuhan Natl Lab Optoelect, Wuhan 430000, Peoples R China

来源：

CHINESE JOURNAL OF CHEMICAL PHYSICS | 2025年 / 38卷 / 01期

基金：

中国国家自然科学基金;

关键词：

Quantum chemistry; Self-consistent field; Hartree-Fock; Electron repulsion integrals; Heterogenous parallel computing; C86/deep computing unit; GRAPHICAL PROCESSING UNITS; EXTENDED HUCKEL THEORY; DENSITY-FUNCTIONAL THEORY; QUANTUM-CHEMISTRY CALCULATIONS; REPULSION INTEGRAL EVALUATION; CONFIGURATION-INTERACTION; MASSIVELY-PARALLEL; HIGH-PERFORMANCE; GPU; PRECISION;

D O I：

10.1063/1674-0068/cjcp2403028

中图分类号：

O64 [物理化学（理论化学）、化学物理学]; O56 [分子物理学、原子物理学];

学科分类号：

070203 ; 070304 ; 081704 ; 1406 ;

摘要：

In this study, we investigate the efficacy of a hybrid parallel algorithm aiming at enhancing the speed of evaluation of two-electron repulsion integrals (ERI) and Fock matrix generation on the Hygon C86/DCU (deep computing unit) heterogeneous computing platform. Multiple hybrid parallel schemes are assessed using a range of model systems, including those with up to 1200 atoms and 10000 basis functions. The findings of our research reveal that, during Hartree-Fock (HF) calculations, a single DCU exhibits 33.6 speedups over 32 C86 CPU cores. Compared with the efficiency of Wuhan Electronic Structure Package on Intel X86 and NVIDIA A100 computing platform, the Hygon platform exhibits good cost-effectiveness, showing great potential in quantum chemistry calculation and other high-performance scientific computations.

引用

页码：81 / 94

页数：14

共 135 条

[1]

AHMADI GR, 1995, CHEM PHYS LETT, V246, P364, DOI 10.1016/0009-2614(95)01127-4

[2] Real-Space Density Functional Theory on Graphical Processing Units: Computational Approach and Comparison to Gaussian Basis Set Methods [J].

Andrade, Xavier ;

Aspuru-Guzik, Alan .

JOURNAL OF CHEMICAL THEORY AND COMPUTATION, 2013, 9 (10) :4360-4373

[3]

Martínez-Martínez LA, 2017, J MEX CHEM SOC, V61, P60

[4] High-Performance Evaluation of High Angular Momentum 4-Center Gaussian Integrals on Modern Accelerated Processors [J].

Asadchev, Andrey ;

Valeev, Edward F. .

JOURNAL OF PHYSICAL CHEMISTRY A, 2023, 127 (51) :10889-10895

[5] Memory-Efficient Recursive Evaluation of 3-Center Gaussian Integrals [J].

Asadchev, Andrey ;

Valeev, Edward F. .

JOURNAL OF CHEMICAL THEORY AND COMPUTATION, 2023, 19 (06) :1698-1710

[6] New Multithreaded Hybrid CPU/GPU Approach to Hartree-Fock [J].

Asadchev, Andrey ;

Gordon, Mark S. .

JOURNAL OF CHEMICAL THEORY AND COMPUTATION, 2012, 8 (11) :4166-4176

[7] Uncontracted Rys Quadrature Implementation of up to G Functions on Graphical Processing Units [J].

Asadchev, Andrey ;

Allada, Veerendra ;

Felder, Jacob ;

Bode, Brett M. ;

Gordon, Mark S. ;

Windus, Theresa L. .

JOURNAL OF CHEMICAL THEORY AND COMPUTATION, 2010, 6 (03) :696-704

[8] Accelerating cluster dynamics simulation of fission gas behavior in nuclear fuel on deep computing unit-based heterogeneous architecture supercomputer [J].

Bai, He ;

Hu, Changjun ;

Zhu, Yuhan ;

Chen, Dandan ;

Chu, Genshen ;

Ren, Shuai .

INTERNATIONAL JOURNAL OF HIGH PERFORMANCE COMPUTING APPLICATIONS, 2023, 37 (05) :516-529

[9] Faster Self-Consistent Field (SCF) Calculations on GPU Clusters [J].

Barca, Giuseppe M. J. ;

Alkan, Melisa ;

Galvez-Vallejo, Jorge L. ;

Poole, David L. ;

Rendell, Alistair P. ;

Gordon, Mark S. .

JOURNAL OF CHEMICAL THEORY AND COMPUTATION, 2021, 17 (12) :7486-7503

[10]

Barca GMJ, 2020, PROCEEDINGS OF SC20: THE INTERNATIONAL CONFERENCE FOR HIGH PERFORMANCE COMPUTING, NETWORKING, STORAGE AND ANALYSIS (SC20)

← 1 2 3 4 5 6 7 8 9 10 →