Large Scale Quantum Chemistry with Tensor Processing Units

被引：19

作者：

Pederson, Ryan ^{[2
,3
,9
]}

Kozlowski, John ^{[1
,2
,3
]}

Song, Ruyi ^{[2
,3
,4
]}

Beall, Jackson ^{[3
,5
]}

Ganahl, Martin ^{[3
]}

Hauru, Markus ^{[3
,6
]}

Lewis, Adam G. M. ^{[3
,5
]}

Yao, Yi ^{[7
]}

Mallick, Shrestha Basu ^{[2
,3
]}

Blum, Volker ^{[4
,7
]}

Vidal, Guifre ^{[2
,3
,8
]}

机构：

[1] Univ Calif Irvine, Dept Chem, Irvine, CA 92617 USA

[2] X Moonshot Factory, Mountain View, CA 94043 USA

[3] SandboxAlphabet, Mountain View, CA 94043 USA

[4] Duke Univ, Dept Chem, Durham, NC 27708 USA

[5] SandboxAQ, Palo Alto, CA 94304 USA

[6] Alan Turing Inst, London NW1 2DB, England

[7] Duke Univ, Thomas Lord Dept Mech Engn & Mat Sci, Durham, NC 27708 USA

[8] Google Quantum AI, Mountain View, CA 94043 USA

[9] Univ Calif Irvine, Dept Phys & Astron, Irvine, CA 92617 USA

来源：

JOURNAL OF CHEMICAL THEORY AND COMPUTATION | 2023年 / 19卷 / 01期

基金：

美国国家科学基金会;

关键词：

DENSITY-FUNCTIONAL CALCULATIONS; PERFORMANCE; ORIGINS;

D O I：

10.1021/acs.jctc.2c00876

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 demonstrate the use of Googles cloud-based Tensor Processing Units (TPUs) to accelerate and scale up conventional (cubic scaling) density functional theory (DFT) calculations. Utilizing 512 TPU cores, we accomplish the largest such DFT computation to date, with 247848 orbitals, corresponding to a cluster of 10327 water molecules with 103270 electrons, all treated explicitly. Our work thus paves the way toward accessible and systematic use of conventional DFT, free of any system-specific constraints, at unprecedented scales.

引用

页码：25 / 32

页数：8

共 50 条

[1] Large-scale distributed linear algebra with tensor processing units
Lewis, Adam G. M.
Beall, Jackson
Ganahl, Martin
Hauru, Markus
Mallick, Shrestha Basu
Vidal, Guifre
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA, 2022, 119 (33)
[2] Extending excited state quantum chemistry to large-scale systems using graphical processing units
Fales, Bryan
Levine, Benjamin
ABSTRACTS OF PAPERS OF THE AMERICAN CHEMICAL SOCIETY, 2016, 252
[3] Graphical Processing Units for Quantum Chemistry
Ufimtsev, Ivan S.
Martinez, Todd J.
COMPUTING IN SCIENCE & ENGINEERING, 2008, 10 (06) : 26 - 34
[4] Multiconfigurational quantum chemistry on graphical processing units
Hohenstein, Edward
ABSTRACTS OF PAPERS OF THE AMERICAN CHEMICAL SOCIETY, 2015, 250
[5] Exploiting graphical processing units (GPUs) to enable large-scale quantum chemistry of solvated molecules with polarizable continuum models
Liu, Fang
Sanchez, David
Ku, Heather
Martinez, Todd
ABSTRACTS OF PAPERS OF THE AMERICAN CHEMICAL SOCIETY, 2018, 256
[6] Accelerating Large-Eddy Simulations of Clouds With Tensor Processing Units
Chammas, Sheide
Wang, Qing
Schneider, Tapio
Ihme, Matthias
Chen, Yi-fan
Anderson, John
JOURNAL OF ADVANCES IN MODELING EARTH SYSTEMS, 2023, 15 (10)
[7] Graphics processing units (GPU) for correlated quantum chemistry calculations
Aspuru-Guzik, Alan
ABSTRACTS OF PAPERS OF THE AMERICAN CHEMICAL SOCIETY, 2010, 240
[8] Large-scale ferrofluid simulations on graphics processing units
Polyakov, A. Yu.
Lyutyy, T. V.
Denisov, S.
Reva, V. V.
Haenggi, P.
COMPUTER PHYSICS COMMUNICATIONS, 2013, 184 (06) : 1483 - 1489
[9] Accelerating correlated quantum chemistry calculations using graphical processing units
Watson, Mark A.
Olivares-Amaya, Roberto
Edgar, Richard G.
Vogt, Leslie
Aspuru-Guzik, Alan
ABSTRACTS OF PAPERS OF THE AMERICAN CHEMICAL SOCIETY, 2010, 240
[10] Accelerating Correlated Quantum Chemistry Calculations Using Graphical Processing Units
Watson, Mark A.
Olivares-Amaya, Roberto
Edgar, Richard G.
Arias, Tomas
Aspuru-Guzik, Alan
COMPUTING IN SCIENCE & ENGINEERING, 2010, 12 (04) : 40 - 50

← 1 2 3 4 5 →