Open Access
Open access
Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2017

An efficient MPI/OpenMP parallelization of the Hartree-Fock method for the second generation of Intel® Xeon Phi™ processor

Alexeev Yuri 2
Keipert Kristopher 3
Dmello Michael 4
Gordon Mark 3
2
 
Argonne National Laboratory
3
 
IOWA STATE UNIVERSITY
4
 
Intel Corporation
5
 
RSC Technologies, Moscow, Russian Federation
Publication typeProceedings Article
Publication date2017-11-12
Abstract
Modern OpenMP threading techniques are used to convert the MPI-only Hartree-Fock code in the GAMESS program to a hybrid MPI/OpenMP algorithm. Two separate implementations that differ by the sharing or replication of key data structures among threads are considered, density and Fock matrices. All implementations are benchmarked on a super-computer of 3,000 Intel® Xeon Phi™ processors. With 64 cores per processor, scaling numbers are reported on up to 192,000 cores. The hybrid MPI/OpenMP implementation reduces the memory footprint by approximately 200 times compared to the legacy code. The MPI/OpenMP code was shown to run up to six times faster than the original for a range of molecular system sizes.

Citations by journals

1
GPS Solutions
GPS Solutions, 1, 12.5%
GPS Solutions
1 publication, 12.5%
Journal of Physical Chemistry A
Journal of Physical Chemistry A, 1, 12.5%
Journal of Physical Chemistry A
1 publication, 12.5%
Journal of Chemical Theory and Computation
Journal of Chemical Theory and Computation, 1, 12.5%
Journal of Chemical Theory and Computation
1 publication, 12.5%
Lecture Notes in Computer Science
Lecture Notes in Computer Science, 1, 12.5%
Lecture Notes in Computer Science
1 publication, 12.5%
1

Citations by publishers

1
2
Springer Nature
Springer Nature, 2, 25%
Springer Nature
2 publications, 25%
American Chemical Society (ACS)
American Chemical Society (ACS), 2, 25%
American Chemical Society (ACS)
2 publications, 25%
1
2
  • We do not take into account publications that without a DOI.
  • Statistics recalculated only for publications connected to researchers, organizations and labs registered on the platform.
  • Statistics recalculated weekly.
Metrics
Share
Cite this
GOST |
Cite this
GOST Copy
Mironov V. A. et al. An efficient MPI/OpenMP parallelization of the Hartree-Fock method for the second generation of Intel® Xeon Phi™ processor // Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2017. 2017.
GOST all authors (up to 50) Copy
Mironov V. A., Alexeev Y., Keipert K., Dmello M., Moskovsky A., Gordon M. An efficient MPI/OpenMP parallelization of the Hartree-Fock method for the second generation of Intel® Xeon Phi™ processor // Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2017. 2017.
RIS |
Cite this
RIS Copy
TY - CPAPER
DO - 10.1145/3126908.3126956
UR - https://doi.org/10.1145%2F3126908.3126956
TI - An efficient MPI/OpenMP parallelization of the Hartree-Fock method for the second generation of Intel® Xeon Phi™ processor
T2 - Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2017
AU - Mironov, Vladimir A.
AU - Alexeev, Yuri
AU - Keipert, Kristopher
AU - Dmello, Michael
AU - Moskovsky, Alexander
AU - Gordon, Mark
PY - 2017
DA - 2017/11/12 00:00:00
ER -
BibTex
Cite this
BibTex Copy
@inproceedings{2017_Mironov
author = {Vladimir A. Mironov and Yuri Alexeev and Kristopher Keipert and Michael Dmello and Alexander Moskovsky and Mark Gordon},
title = {An efficient MPI/OpenMP parallelization of the Hartree-Fock method for the second generation of Intel® Xeon Phi™ processor},
year = {2017},
month = {nov}
}
Found error?