Spatial localization of high-frequency particle oscillations in a complex plasma monolayer

IV Voronov, VS Nikolaev, AV Timofeev… - Journal of Physics D …, 2023 - iopscience.iop.org
We study the spatial distribution of dynamical properties in a strongly coupled confined
experimental dust monolayer with a dense central region and a rarefied periphery. We show …

GPU-based molecular dynamics of fluid flows: Reaching for turbulence

D Pavlov, V Galigerov, D Kolotinskii… - … Journal of High …, 2024 - journals.sagepub.com
Fluid dynamics is a ubiquitous problem that arises in different branches of science and
industry. It is usually tackled by numerically solving continuum Navier-Stokes type …

Graph based routing algorithm for torus topology and its evaluation for the Angara interconnect

A Mukosey, A Semenov, A Tretiakov - Journal of Parallel and Distributed …, 2024 - Elsevier
Several approaches and techniques exist to resolve load balancing problem in general and
torus topology networks. Graph methods are natural ways to perform balancing of routing …

An efficient sparse stiffness matrix vector multiplication using compressed sparse row storage format on AMD GPU

L **ng, Z Wang, Z Ding, G Chu… - Concurrency and …, 2022 - Wiley Online Library
The performance of sparse stiffness matrix‐vector multiplication is essential for large‐scale
structural mechanics numerical simulation. Compressed sparse row (CSR) is the most …

Towards OpenUCX and GPUDirect technology support for the Angara interconnect

M Khalilov, A Timofeev, D Polyakov - Russian Supercomputing Days, 2022 - Springer
Modern supercomputers consist of thousands of GPUs interconnected by high-speed HPC
fabric. Angara interconnect supports Remote Direct Memory Access (RDMA) semantic …

GPU-based molecular dynamics of turbulent liquid flows with OpenMM

D Pavlov, D Kolotinskii, V Stegailov - International Conference on Parallel …, 2022 - Springer
In this paper we describe the computational framework for GPU-based molecular dynamics
of turbulent flows. The framework is based on the open-source molecular dynamics library …

The High Performance Interconnect Architecture for Supercomputers

AS Simonov, AS Semenov, AN Shcherbak… - Supercomputing …, 2023 - superfri.susu.ru
In this paper, we introduce the design of an advanced high-performance interconnect
architecture for supercomputers. In the first part of the paper, we consider the first generation …

Early Performance Evaluation of Distributed Learning via GPU-Aware MPI over the Angara Interconnect

TF Ismagilov, AV Piruygin, AS Semenov - Lobachevskii Journal of …, 2023 - Springer
Distributed neural network computation heavily relies on GPUs and high-speed
interconnects. We use Nvidia Tesla GPUs with Angara interconnect for distributed deep …

Collective motions of atoms in crystals

VD Negodin, DI Fleita, GE Norman - Mathematical Models and Computer …, 2023 - Springer
The collective motions of atoms in the molecular dynamics model of a classical Lennard-
Jones monocrystal are studied in the process of heating up to a spinodal decomposition and …

Optimized graph based routing algorithm for the Angara interconnect

A Mukosey, A Semenov, A Tretiakov - arxiv preprint arxiv:2110.00851, 2021 - arxiv.org
JSC NICEVT has developed the Angara high-speed interconnect with 4D torus topology.
The Angara interconnect router implements deterministic routing based on the bubble flow …