本文へ移動

サイトナビゲーションへ移動

検索ボックスへ移動

サイドバーへ移動

ここは、本文エリアの先頭です。

Development of a Many-block Method for Enabling an Extremely Fast Combustion LES

JAXA Supercomputer System Annual Report April 2018-March 2019

Report Number: R18EDA201N13

Subject Category: Aeronautical Technology

PDF available here

  • Responsible Representative: Shingo Matsuyama, Aeronautical Technology Directorate, Numerical Simulation Research Unit
  • Contact Information: Shingo Matsuyama(smatsu@chofu.jaxa.jp)
  • Members: Shingo Matsuyama

Abstract

In this research, by the following two approaches: (1) acceleration of time integration by implicit time integration method and (2) acceleration of calculation using an accelerator, we try to achieve a speedup of more than 100 times compared with the current level of CFD runs. We aim to develop a “many-block method” that can complete a single case of turbulent combustion LES in an extremely short time period.

Reference URL

N/A

Reasons for using JSS2

We used JSS2 for developing a LES code by a “many-block method” that has inner cells for high-order interpolation inside an unstructured cell, and for performing LES analysis with 100 million grid points.

Achievements of the Year

In this research, we developed a LES code by “many-block method” that has inner cells for high-order interpolation inside an unstructured cell. We performed an LES analysis for a cylinder at Re = 3900 with about 63 million unstrectured hexa-cells (Figs.1 and 2).

Annual Reoprt Figures for 2018

Fig.1: Example of unstructured hexa-cells around a cylinder. A single hexa-cell has 5x5x5 subcells in its inside. Cited from non peer-reviewed paper [2].

 

Annual Reoprt Figures for 2018

Fig.2: Instantaneous contours of vorticity magnitude on the x-y plane for the LES around a cylinder at Re = 3900. Cited from non peer-reviewed paper [2].

 

Publications

– Non peer-reviewed papers

1) Shingo Matsuyama, “High-order Unstructured-grid Large Eddy Simulation Using Many-block Method”, Proceedings of the 50th Fluid Dynamics Conference /

the 36th Aerospace Numerical Simulation Symposium, 3E02, 2018.

2) Shingo Matsuyama, “Unstructured-grid LES Using Many-block Method”, Proceedings of the 32nd CFD Sympoium, C08-3, 2018.

– Oral Presentations

1) Shingo Matsuyama, “High-order Unstructured-grid Large Eddy Simulation Using Many-block Method”, the 50th Fluid Dynamics Conference /

the 36th Aerospace Numerical Simulation Symposium, 2018.

2) Shingo Matsuyama, “Unstructured-grid LES Using Many-block Method”, the 32nd CFD Sympoium, 2018.

Usage of JSS2

Computational Information

  • Process Parallelization Methods: MPI
  • Thread Parallelization Methods: OpenMP
  • Number of Processes: 64 – 256
  • Elapsed Time per Case: 48 Hour(s)

Resources Used

 

Fraction of Usage in Total Resources*1(%): 0.00

 

Details

Please refer to System Configuration of JSS2 for the system configuration and major specifications of JSS2.

Computational Resources
System Name Amount of Core Time
(core x hours)
Fraction of Usage*2(%)
SORA-MA 5,106.17 0.00
SORA-PP 0.00 0.00
SORA-LM 0.00 0.00
SORA-TPP 0.00 0.00

 

File System Resources
File System Name Storage Assigned
(GiB)
Fraction of Usage*2(%)
/home 399.29 0.41
/data 1,899.35 0.03
/ltmp 325.52 0.03

 

Archiver Resources
Archiver Name Storage Used
(TiB)
Fraction of Usage*2(%)
J-SPACE 0.00 0.00

*1: Fraction of Usage in Total Resources: Weighted average of three resource types (Computing, File System, and Archiver).

*2: Fraction of Usage:Percentage of usage relative to each resource used in one year.

JAXA Supercomputer System Annual Report April 2018-March 2019