Related Experiment Video
Updated: Oct 27, 2025

Operation of the Collaborative Composite Manufacturing CCM System
Published on: October 1, 2019
A Massively Parallel Implementation of the CCSD(T) Method Using the Resolution-of-the-Identity Approximation and a
Dipayan Datta1, Mark S Gordon1
1Department of Chemistry and Ames Laboratory, Iowa State University, 2416 Pammel Drive, Ames 50011-2416, Iowa United States of America.
This study introduces a new parallel algorithm for coupled-cluster singles and doubles with perturbative triples [CCSD(T)] calculations, improving efficiency for large molecular systems. The algorithm achieves near-linear scaling, enabling accurate interaction energy calculations for complex molecules.
Area of Science:
- Computational Chemistry
- Quantum Chemistry
- High-Performance Computing
Background:
- Coupled-cluster singles and doubles with perturbative triples [CCSD(T)] is a high-accuracy quantum chemistry method.
- Large-scale CCSD(T) calculations are computationally demanding due to significant memory and CPU requirements.
- The resolution-of-the-identity (RI) approximation is crucial for reducing the computational cost of calculating two-electron repulsion integrals (ERIs).
Purpose of the Study:
- To develop and implement a parallel algorithm for the RI-approximated CCSD(T) method.
- To optimize parallel efficiency for both the CCSD amplitude equations and the computationally intensive triples correction.
- To enable accurate electronic structure calculations for larger and more complex molecular systems.
Main Methods:
- An integral-direct strategy is employed to bypass the storage of four-center ERIs.
- A hybrid MPI/OpenMP parallelization scheme is utilized for efficient intranode and internode communication.
- Two algorithms are implemented for rate-limiting terms, differing in memory usage and scaling behavior.
- Near-linear scaling is achieved for the CCSD(T) calculation on up to 8000 cores.
Main Results:
- The developed parallel algorithm demonstrates near-linear scaling, particularly for the triples correction step.
- Two distinct algorithms for rate-limiting terms show trade-offs between speed and memory efficiency at different core counts.
- Successful application to molecules with up to 51 atoms and 1624 basis functions.
- Accurate calculation of the complete basis set (CBS) limit for the interaction energy of the uracil dimer.
Conclusions:
- The implemented parallel RI-CCSD(T) algorithm significantly enhances computational efficiency for high-accuracy electronic structure calculations.
- The algorithm's scalability allows for routine application to larger molecular systems.
- The accurate interaction energy calculation for the uracil dimer validates the method's performance and applicability.
More Related Videos
10:52Multiscale Sampling of a Heterogeneous Water/Metal Catalyst Interface using Density Functional Theory and Force-Field Molecular Dynamics
Published on: April 12, 2019
07:52Author Spotlight: Enhancing Cryo-Electron Microscopy by Automated Data Collection and Analysis Techniques
Published on: December 1, 2023
Related Concept Videos
Parallel Processing
Distributed Loads: Problem Solving
Parallel Resonance
Ampere-Maxwell's Law: Problem-Solving
To solve the problem, we can use the equations from the analysis of an RC circuit and Maxwell's version of Ampère's law.
For the first part of...
Mechanistic Models: Compartment Models in Algorithms for Numerical Problem Solving
In individual population analyses, different algorithms are employed, such as Cauchy's method, which uses a...
Parallel-axis Theorem