volume 33 issue 3 pages 507-521

Angara interconnect makes GPU-based Desmos supercomputer an efficient tool for molecular dynamics calculations

Ekaterina Dlinnova 2
Timur Ismagilov 3
Mikhail Khalilov 2
Dmitry Makagon 3
Alexander Semenov 2, 3
Alexei Simonov 3
Grigory Smirnov 1, 2
Publication typeJournal Article
Publication date2019-02-20
scimago Q1
wos Q2
SJR1.051
CiteScore7.4
Impact factor2.5
ISSN10943420, 17412846
Hardware and Architecture
Software
Theoretical Computer Science
Abstract

In this article, we describe the Desmos supercomputer that consists of 32 hybrid nodes connected by a low-latency high-bandwidth Angara interconnect with torus topology. This supercomputer is aimed at cost-effective classical molecular dynamics calculations. Desmos serves as a test bed for the Angara interconnect that supports 3-D and 4-D torus network topologies and verifies its ability to unite massively parallel programming systems speeding-up effectively message-passing interface (MPI)-based applications. We describe the Angara interconnect presenting typical MPI benchmarks. Desmos benchmarks results for GROMACS, LAMMPS, VASP and CP2K are compared with the data for other high-performance computing (HPC) systems. Also, we consider the job scheduling statistics for several months of Desmos deployment.

Found 
Found 

Top-30

Journals

2
4
6
8
10
Lecture Notes in Computer Science
10 publications, 18.52%
Journal of Molecular Liquids
6 publications, 11.11%
Journal of Chemical Physics
5 publications, 9.26%
Communications in Computer and Information Science
4 publications, 7.41%
Fluid Phase Equilibria
3 publications, 5.56%
International Journal of High Performance Computing Applications
3 publications, 5.56%
JETP Letters
2 publications, 3.7%
Lobachevskii Journal of Mathematics
2 publications, 3.7%
Crystal Growth and Design
1 publication, 1.85%
Journal of the Brazilian Society of Mechanical Sciences and Engineering
1 publication, 1.85%
Computational Materials Science
1 publication, 1.85%
Computer Physics Communications
1 publication, 1.85%
Journal of Physics: Conference Series
1 publication, 1.85%
Journal of Nuclear Materials
1 publication, 1.85%
Polymer Science - Series A
1 publication, 1.85%
Uspekhi Fizicheskih Nauk
1 publication, 1.85%
Matematicheskoe modelirovanie
1 publication, 1.85%
Polymer Science - Series C
1 publication, 1.85%
Journal Physics D: Applied Physics
1 publication, 1.85%
CCF Transactions on High Performance Computing
1 publication, 1.85%
Journal of Parallel and Distributed Computing
1 publication, 1.85%
Mathematical Models and Computer Simulations
1 publication, 1.85%
Chinese Journal of Aeronautics
1 publication, 1.85%
Высокомолекулярные соединения С
1 publication, 1.85%
Russian Journal of Physical Chemistry A
1 publication, 1.85%
2
4
6
8
10

Publishers

2
4
6
8
10
12
14
16
Springer Nature
16 publications, 29.63%
Elsevier
14 publications, 25.93%
Pleiades Publishing
8 publications, 14.81%
AIP Publishing
5 publications, 9.26%
SAGE
3 publications, 5.56%
Institute of Electrical and Electronics Engineers (IEEE)
2 publications, 3.7%
IOP Publishing
2 publications, 3.7%
American Chemical Society (ACS)
1 publication, 1.85%
Uspekhi Fizicheskikh Nauk Journal
1 publication, 1.85%
Keldysh Institute of Applied Mathematics
1 publication, 1.85%
The Russian Academy of Sciences
1 publication, 1.85%
2
4
6
8
10
12
14
16
  • We do not take into account publications without a DOI.
  • Statistics recalculated weekly.

Are you a researcher?

Create a profile to get free access to personal recommendations for colleagues and new articles.
Metrics
54
Share
Cite this
GOST |
Cite this
GOST Copy
Stegailov V. et al. Angara interconnect makes GPU-based Desmos supercomputer an efficient tool for molecular dynamics calculations // International Journal of High Performance Computing Applications. 2019. Vol. 33. No. 3. pp. 507-521.
GOST all authors (up to 50) Copy
Stegailov V., Dlinnova E., Ismagilov T., Khalilov M., Kondratyuk N., Makagon D., Semenov A., Simonov A., Smirnov G., Timofeev A. Angara interconnect makes GPU-based Desmos supercomputer an efficient tool for molecular dynamics calculations // International Journal of High Performance Computing Applications. 2019. Vol. 33. No. 3. pp. 507-521.
RIS |
Cite this
RIS Copy
TY - JOUR
DO - 10.1177/1094342019826667
UR - https://journals.sagepub.com/doi/10.1177/1094342019826667
TI - Angara interconnect makes GPU-based Desmos supercomputer an efficient tool for molecular dynamics calculations
T2 - International Journal of High Performance Computing Applications
AU - Stegailov, Vladimir V.
AU - Dlinnova, Ekaterina
AU - Ismagilov, Timur
AU - Khalilov, Mikhail
AU - Kondratyuk, Nikolay
AU - Makagon, Dmitry
AU - Semenov, Alexander
AU - Simonov, Alexei
AU - Smirnov, Grigory
AU - Timofeev, Alexey
PY - 2019
DA - 2019/02/20
PB - SAGE
SP - 507-521
IS - 3
VL - 33
SN - 1094-3420
SN - 1741-2846
ER -
BibTex |
Cite this
BibTex (up to 50 authors) Copy
@article{2019_Stegailov,
author = {Vladimir V. Stegailov and Ekaterina Dlinnova and Timur Ismagilov and Mikhail Khalilov and Nikolay Kondratyuk and Dmitry Makagon and Alexander Semenov and Alexei Simonov and Grigory Smirnov and Alexey Timofeev},
title = {Angara interconnect makes GPU-based Desmos supercomputer an efficient tool for molecular dynamics calculations},
journal = {International Journal of High Performance Computing Applications},
year = {2019},
volume = {33},
publisher = {SAGE},
month = {feb},
url = {https://journals.sagepub.com/doi/10.1177/1094342019826667},
number = {3},
pages = {507--521},
doi = {10.1177/1094342019826667}
}
MLA
Cite this
MLA Copy
Stegailov, Vladimir V., et al. “Angara interconnect makes GPU-based Desmos supercomputer an efficient tool for molecular dynamics calculations.” International Journal of High Performance Computing Applications, vol. 33, no. 3, Feb. 2019, pp. 507-521. https://journals.sagepub.com/doi/10.1177/1094342019826667.