INQ, a Modern GPU-Accelerated Computational Framework for (Time-Dependent) Density Functional Theory

被引：15

作者：

Andrade, Xavier ^{[3
]}

Das Pemmaraju, Chaitanya ^{[1
]}

Kartsev, Alexey ^{[1
]}

Xiao, Jun ^{[1
]}

Lindenberg, Aaron ^{[1
]}

Rajpurohit, Sangeeta ^{[2
]}

Tan, Liang Z. ^{[2
]}

Ogitsu, Tadashi ^{[3
]}

Correa, Alfredo A. ^{[3
]}

机构：

[1] SLAC Natl Accelerator Lab, Stanford Inst Mat & Energy Sci, Menlo Pk, CA 94025 USA

[2] Lawrence Berkeley Natl Lab, Mol Foundry, Berkeley, CA 94720 USA

[3] Lawrence Livermore Natl Lab, Quantum Simulat Grp, Livermore, CA 94551 USA

来源：

JOURNAL OF CHEMICAL THEORY AND COMPUTATION | 2021年 / 17卷 / 12期

关键词：

ELECTRONIC-STRUCTURE CALCULATIONS; TOTAL-ENERGY CALCULATIONS; MOLECULAR-DYNAMICS; OPTICAL-RESPONSE; PSEUDOPOTENTIALS; PERFORMANCE; EXCHANGE; EXCITATIONS; SCATTERING; CHEMISTRY;

D O I：

10.1021/acs.jctc.1c00562

中图分类号：

O64 [物理化学（理论化学）、化学物理学];

学科分类号：

070304 ; 081704 ;

摘要：

We present INQ, a new implementation of density functional theory (DFT) and time-dependent DFT (TDDFT) written from scratch to work on graphic processing units (GPUs). Besides GPU support, INQ makes use of modern code design features and takes advantage of newly available hardware. By designing the code around algorithms, rather than against specific implementations and numerical libraries, we aim to provide a concise and modular code. The result is a fairly complete DFT/TDDFT implementation in roughly 12 000 lines of open-source C++ code representing a modular platform for community-driven application development on emerging high-performance computing architectures.

引用

页码：7447 / 7467

页数：21

共 194 条

[31] BYCHKOV YA, 1984, JETP LETT+, V39, P78
[32] Caporaso GJ, 2009, REV ACCEL SCI TECH, V2, P253, DOI 10.1142/S1793626809000235
[33] Fast hybrid density-functional computations using plane-wave basis sets
Carnimeo, Ivan
Baroni, Stefano
Giannozzi, Paolo
[J]. ELECTRONIC STRUCTURE, 2019, 1 (01):
[34] Propagators for the time-dependent Kohn-Sham equations
Castro, A
Marques, MAL
Rubio, A
[J]. JOURNAL OF CHEMICAL PHYSICS, 2004, 121 (08) : 3425 - 3433
[35] octopus: a tool for the application of time-dependent density functional theory
Castro, Alberto
Appel, Heiko
Oliveira, Micael
Rozzi, Carlo A.
Andrade, Xavier
Lorenzen, Florian
Marques, M. A. L.
Gross, E. K. U.
Rubio, Angel
[J]. PHYSICA STATUS SOLIDI B-BASIC SOLID STATE PHYSICS, 2006, 243 (11): : 2465 - 2488
[36] The challenge of predicting optical properties of biomolecules: What can we learn from time-dependent density-functional theory?
Castro, Alberto
Marques, Miguel A. L.
Varsano, Daniele
Sottile, Francesco
Rubio, Angel
[J]. COMPTES RENDUS PHYSIQUE, 2009, 10 (06) : 469 - 490
[37] Photoinduced melting of magnetic order in the correlated electron insulator NdNiO3
Caviglia, A. D.
Foerst, M.
Scherwitzl, R.
Khanna, V.
Bromberger, H.
Mankowsky, R.
Singla, R.
Chuang, Y. -D.
Lee, W. S.
Krupin, O.
Schlotter, W. F.
Turner, J. J.
Dakovski, G. L.
Minitti, M. P.
Robinson, J.
Scagnoli, V.
Wilkins, S. B.
Cavill, S. A.
Gibert, M.
Gariglio, S.
Zubko, P.
Triscone, J. -M.
Hill, J. P.
Dhesi, S. S.
Cavalleri, A.
[J]. PHYSICAL REVIEW B, 2013, 88 (22):
[38] FINITE-DIFFERENCE-PSEUDOPOTENTIAL METHOD - ELECTRONIC-STRUCTURE CALCULATIONS WITHOUT A BASIS
CHELIKOWSKY, JR
TROULLIER, N
SAAD, Y
[J]. PHYSICAL REVIEW LETTERS, 1994, 72 (08) : 1240 - 1243
[39] Performance Evaluation of Advanced Features in CUDA Unified Memory
Chien, Steven W. D.
Peng, Ivy B.
Markidis, Stefano
[J]. PROCEEDINGS OF MCHPC'19: 2019 IEEE/ACM WORKSHOP ON MEMORY CENTRIC HIGH PERFORMANCE COMPUTING (MCHPC), 2019, : 50 - 57
[40] Insights into current limitations of density functional theory
Cohen, Aron J.
Mori-Sanchez, Paula
Yang, Weitao
[J]. SCIENCE, 2008, 321 (5890) : 792 - 794

← 1 2 3 4 5 6 7 8 9 10 →