Search engine for discovering works of Art, research articles, and books related to Art and Culture
Javascript must be enabled to continue!
Large-Scale Heterogeneous Computing for 3D Deterministic Particle Transport on Tianhe-2A Supercomputer
View through CrossRef
Scalable parallel algorithm for particle transport is one of the main application fields in high-performance computing. Discrete ordinate method (Sn) is one of the most popular deterministic numerical methods for solving particle transport equations. In this paper, we introduce a new method of large-scale heterogeneous computing of one energy group time-independent deterministic discrete ordinates neutron transport in 3D Cartesian geometry (Sweep3D) on Tianhe-2A supercomputer. In heterogeneous programming, we use customized Basic Communication Library (BCL) and Accelerated Computing Library (ACL) to control and communicate between CPU and the Matrix2000 accelerator. We use OpenMP instructions to exploit the parallelism of threads based on Matrix 2000. The test results show that the optimization of applying OpenMP on particle transport algorithm modified by our method can get 11.3 times acceleration at most. On Tianhe-2A supercomputer, the parallel efficiency of 1.01 million cores compared with 170 thousand cores is 52%.
Title: Large-Scale Heterogeneous Computing for 3D Deterministic Particle Transport on Tianhe-2A Supercomputer
Description:
Scalable parallel algorithm for particle transport is one of the main application fields in high-performance computing.
Discrete ordinate method (Sn) is one of the most popular deterministic numerical methods for solving particle transport equations.
In this paper, we introduce a new method of large-scale heterogeneous computing of one energy group time-independent deterministic discrete ordinates neutron transport in 3D Cartesian geometry (Sweep3D) on Tianhe-2A supercomputer.
In heterogeneous programming, we use customized Basic Communication Library (BCL) and Accelerated Computing Library (ACL) to control and communicate between CPU and the Matrix2000 accelerator.
We use OpenMP instructions to exploit the parallelism of threads based on Matrix 2000.
The test results show that the optimization of applying OpenMP on particle transport algorithm modified by our method can get 11.
3 times acceleration at most.
On Tianhe-2A supercomputer, the parallel efficiency of 1.
01 million cores compared with 170 thousand cores is 52%.
Related Results
A simplified Python-based kinematic model of particle transport in rivers
A simplified Python-based kinematic model of particle transport in rivers
We present results from a particle-scale numerical model inspired by the idea that a majority of the time during transport capable floods, bedload transport in rivers is rarefied, ...
Experimental and numerical investigation into the effect of surface roughness on particle rebound
Experimental and numerical investigation into the effect of surface roughness on particle rebound
Erosion damage and particle deposition are crucial wear phenomena in gas turbine engines. As a result, compressor efficiency decreases, stability margin reduces, and maintenance co...
Deploying and scaling distributed parallel deep neural networks on the Tianhe-3 prototype system
Deploying and scaling distributed parallel deep neural networks on the Tianhe-3 prototype system
AbstractDue to the increase in computing power, it is possible to improve the feature extraction and data fitting capabilities of DNN networks by increasing their depth and model c...
SYSTEMATIZATION OF THE REGULATORY FRAMEWORK OF ENSURING THE WATER TRANSPORT COMPETITIVENESS IN UKRAINE
SYSTEMATIZATION OF THE REGULATORY FRAMEWORK OF ENSURING THE WATER TRANSPORT COMPETITIVENESS IN UKRAINE
Topicality. Business entities in the field of water transport can gain competitive advantages and ensure their competitiveness through the introduction of innovations into the proc...
GUIDING PRINCIPLES OF PRIORITIZING TASKS FOR DEVELOPING UKRAINE'S WATER TRANSPORT
GUIDING PRINCIPLES OF PRIORITIZING TASKS FOR DEVELOPING UKRAINE'S WATER TRANSPORT
Topicality. Water transport is of crucial importance for the economic prosperity and national security of Ukraine. Most sectors of the economy depend on water transport and its inf...
Parametric study of the shock driven multiphase flow
Parametric study of the shock driven multiphase flow
This thesis presents a study of the shock driven multiphase instability and explores the competing effects of particle and gas effects in mixing. Eight different simulations were c...
Asteroid Regolith Strength Modelling Informed by Bennu Sample Measurements
Asteroid Regolith Strength Modelling Informed by Bennu Sample Measurements
Abstract Informed by findings about the sample from asteroid Bennu, we apply a new simulation methodology to better understand how Van der Waals (VdW) forces work between spherical...
Diffusion in heterogeneous discs and spheres: New closed-form expressions for exit times and homogenization formulas
Diffusion in heterogeneous discs and spheres: New closed-form expressions for exit times and homogenization formulas
Mathematical models of diffusive transport underpin our understanding of chemical, biochemical, and biological transport phenomena. Analysis of such models often focuses on relativ...