DCA++: A software framework to solve correlated electron problems with modern quantum cluster methods

Urs R. Hähner; Gonzalo Alvarez; Thomas A. Maier; Raffaele Solcà; Peter Staar; Michael S. Summers; Thomas C. Schulthess

doi:10.1016/j.cpc.2019.01.006

DCA++: A software framework to solve correlated electron problems with modern quantum cluster methods

Urs R. Hähner, Gonzalo Alvarez, Thomas A. Maier, Raffaele Solcà, Peter Staar, Michael S. Summers, Thomas C. Schulthess

Computational Sciences & Engineering Div

Research output: Contribution to journal › Article › peer-review

18 Scopus citations

Abstract

We present the first open release of the DCA++ project, a high-performance research software framework to solve quantum many-body problems with cutting edge quantum cluster algorithms. DCA++ implements the dynamical cluster approximation (DCA) and its DCA⁺ extension with a continuous self-energy. The algorithms capture nonlocal correlations in strongly correlated electron systems, thereby giving insight into high-T_c superconductivity. The code's scalability allows efficient usage of systems at all scales, from workstations to leadership computers. With regard to the increasing heterogeneity of modern computing machines, DCA++ provides portable performance on conventional and emerging new architectures, such as hybrid CPU–GPU, sustaining multiple petaflops on ORNL's Titan and CSCS’ Piz Daint supercomputers. Moreover, we show how sustainable and scalable development of the code base has been achieved by adopting standard techniques of the software industry. These include employing a distributed version control system, applying test-driven development and following continuous integration. Program summary: Program Title: DCA++ Program Files doi: http://dx.doi.org/10.17632/482jm5cv77.1 Licensing provisions: BSD-3-Clause Programming language: C++14 and CUDA Nature of problem: Understanding the fascinating physics of strongly correlated electron systems requires the development of sophisticated algorithms and their implementation on leadership computing systems. Solution method: The DCA++ code provides a highly scalable and efficient implementation of the dynamical cluster approximation (DCA) and its DCA⁺ extension.

Original language	English
Article number	106709
Journal	Computer Physics Communications
Volume	246
DOIs	https://doi.org/10.1016/j.cpc.2019.01.006
State	Published - Jan 2020

Funding

The authors would like to acknowledge the contributions of Giovanni Balduzzi, Peter W. Doak, Mi Jiang, Andrei Plamada and Bart Ydens to the DCA++ project. We would like to thank John Biddiscombe for his contributions to refactoring the code base, Guilherme Peretti-Pezzi for his assistance in setting up continuous integration and his support with EasyBuild, and Philipp Werner for providing the CT-HYB QMC solver. U.R.H. acknowledges the support by the National Center of Competence in Research (NCCR) MARVEL, funded by the Swiss National Science Foundation. The work of T.A.M. was supported by the Scientific Discovery through Advanced Computing (SciDAC) program funded by U.S. Department of Energy, Office of Science, Advanced Scientific Computing Research and Basic Energy Sciences, Division of Materials Sciences and Engineering. This research used resources of the Oak Ridge Leadership Computing Facility (OLCF) awarded by the INCITE program, and of the Swiss National Supercomputing Center (CSCS). OLCF is a DOE Office of Science User Facility supported under Contract DE-AC05-00OR22725. The authors would like to thank John Biddiscombe for his contributions to refactoring the code base and Guilherme Peretti-Pezzi for his assistance in setting up continuous integration and his support with EasyBuild. U.R.H. acknowledges the support by the National Center of Competence in Research (NCCR) MARVEL, funded by the Swiss National Science Foundation. The work of T.A.M. was supported by the Scientific Discovery through Advanced Computing (SciDAC) program funded by U.S. Department of Energy, Office of Science, Advanced Scientific Computing Research and Basic Energy Sciences, Division of Materials Sciences and Engineering. This research used resources of the Oak Ridge Leadership Computing Facility (OLCF) awarded by the INCITE program, and of the Swiss National Supercomputing Center (CSCS). OLCF is a DOE Office of Science User Facility supported under Contract DE-AC05-00OR22725.

Funders	Funder number
Swiss National Supercomputing Center
U.S. Department of Energy
Office of Science
Basic Energy Sciences	DE-AC05-00OR22725
Advanced Scientific Computing Research
nccr – on the move
Division of Materials Sciences and Engineering
Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung

Keywords

Continuous-time quantum Monte Carlo
Dynamical cluster approximation
Extreme-scale computing
Quantum cluster algorithms
Strongly correlated electron systems
Sustainable software development

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1016/j.cpc.2019.01.006

Cite this

@article{8abd27a78f304f38b589da055a03c560,

title = "DCA++: A software framework to solve correlated electron problems with modern quantum cluster methods",

abstract = "We present the first open release of the DCA++ project, a high-performance research software framework to solve quantum many-body problems with cutting edge quantum cluster algorithms. DCA++ implements the dynamical cluster approximation (DCA) and its DCA+ extension with a continuous self-energy. The algorithms capture nonlocal correlations in strongly correlated electron systems, thereby giving insight into high-Tc superconductivity. The code's scalability allows efficient usage of systems at all scales, from workstations to leadership computers. With regard to the increasing heterogeneity of modern computing machines, DCA++ provides portable performance on conventional and emerging new architectures, such as hybrid CPU–GPU, sustaining multiple petaflops on ORNL's Titan and CSCS{\textquoteright} Piz Daint supercomputers. Moreover, we show how sustainable and scalable development of the code base has been achieved by adopting standard techniques of the software industry. These include employing a distributed version control system, applying test-driven development and following continuous integration. Program summary: Program Title: DCA++ Program Files doi: http://dx.doi.org/10.17632/482jm5cv77.1 Licensing provisions: BSD-3-Clause Programming language: C++14 and CUDA Nature of problem: Understanding the fascinating physics of strongly correlated electron systems requires the development of sophisticated algorithms and their implementation on leadership computing systems. Solution method: The DCA++ code provides a highly scalable and efficient implementation of the dynamical cluster approximation (DCA) and its DCA+ extension.",

keywords = "Continuous-time quantum Monte Carlo, Dynamical cluster approximation, Extreme-scale computing, Quantum cluster algorithms, Strongly correlated electron systems, Sustainable software development",

author = "H{\"a}hner, {Urs R.} and Gonzalo Alvarez and Maier, {Thomas A.} and Raffaele Solc{\`a} and Peter Staar and Summers, {Michael S.} and Schulthess, {Thomas C.}",

note = "Publisher Copyright: {\textcopyright} 2019 The Authors",

year = "2020",

month = jan,

doi = "10.1016/j.cpc.2019.01.006",

language = "English",

volume = "246",

journal = "Computer Physics Communications",

issn = "0010-4655",

publisher = "Elsevier",

}

TY - JOUR

T1 - DCA++

T2 - A software framework to solve correlated electron problems with modern quantum cluster methods

AU - Hähner, Urs R.

AU - Alvarez, Gonzalo

AU - Maier, Thomas A.

AU - Solcà, Raffaele

AU - Staar, Peter

AU - Summers, Michael S.

AU - Schulthess, Thomas C.

PY - 2020/1

Y1 - 2020/1

N2 - We present the first open release of the DCA++ project, a high-performance research software framework to solve quantum many-body problems with cutting edge quantum cluster algorithms. DCA++ implements the dynamical cluster approximation (DCA) and its DCA+ extension with a continuous self-energy. The algorithms capture nonlocal correlations in strongly correlated electron systems, thereby giving insight into high-Tc superconductivity. The code's scalability allows efficient usage of systems at all scales, from workstations to leadership computers. With regard to the increasing heterogeneity of modern computing machines, DCA++ provides portable performance on conventional and emerging new architectures, such as hybrid CPU–GPU, sustaining multiple petaflops on ORNL's Titan and CSCS’ Piz Daint supercomputers. Moreover, we show how sustainable and scalable development of the code base has been achieved by adopting standard techniques of the software industry. These include employing a distributed version control system, applying test-driven development and following continuous integration. Program summary: Program Title: DCA++ Program Files doi: http://dx.doi.org/10.17632/482jm5cv77.1 Licensing provisions: BSD-3-Clause Programming language: C++14 and CUDA Nature of problem: Understanding the fascinating physics of strongly correlated electron systems requires the development of sophisticated algorithms and their implementation on leadership computing systems. Solution method: The DCA++ code provides a highly scalable and efficient implementation of the dynamical cluster approximation (DCA) and its DCA+ extension.

AB - We present the first open release of the DCA++ project, a high-performance research software framework to solve quantum many-body problems with cutting edge quantum cluster algorithms. DCA++ implements the dynamical cluster approximation (DCA) and its DCA+ extension with a continuous self-energy. The algorithms capture nonlocal correlations in strongly correlated electron systems, thereby giving insight into high-Tc superconductivity. The code's scalability allows efficient usage of systems at all scales, from workstations to leadership computers. With regard to the increasing heterogeneity of modern computing machines, DCA++ provides portable performance on conventional and emerging new architectures, such as hybrid CPU–GPU, sustaining multiple petaflops on ORNL's Titan and CSCS’ Piz Daint supercomputers. Moreover, we show how sustainable and scalable development of the code base has been achieved by adopting standard techniques of the software industry. These include employing a distributed version control system, applying test-driven development and following continuous integration. Program summary: Program Title: DCA++ Program Files doi: http://dx.doi.org/10.17632/482jm5cv77.1 Licensing provisions: BSD-3-Clause Programming language: C++14 and CUDA Nature of problem: Understanding the fascinating physics of strongly correlated electron systems requires the development of sophisticated algorithms and their implementation on leadership computing systems. Solution method: The DCA++ code provides a highly scalable and efficient implementation of the dynamical cluster approximation (DCA) and its DCA+ extension.

KW - Continuous-time quantum Monte Carlo

KW - Dynamical cluster approximation

KW - Extreme-scale computing

KW - Quantum cluster algorithms

KW - Strongly correlated electron systems

KW - Sustainable software development

UR - http://www.scopus.com/inward/record.url?scp=85060536227&partnerID=8YFLogxK

U2 - 10.1016/j.cpc.2019.01.006

DO - 10.1016/j.cpc.2019.01.006

M3 - Article

AN - SCOPUS:85060536227

SN - 0010-4655

VL - 246

JO - Computer Physics Communications

JF - Computer Physics Communications

M1 - 106709

ER -

DCA++: A software framework to solve correlated electron problems with modern quantum cluster methods

Abstract

Funding

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this