JAXA Supercomputer System Annual Report April 2019-March 2020

Report Number: R19ECMP06

Subject Category: Competitive Funding

PDF available here

• Responsible Representative: Yuko Inatomi, Institute of Space and Astronautical Science, Department of Interdisciplinary Space Science
• Contact Information: Ryoji Takaki(ryo@isas.jaxa.jp)
• Members: Ryoji Takaki, Taku Nonomura, Seiji Tsutsumi, Yuma Fukushima, Soshi Kawai, Ikuo Miyoshi, Satoshi Sekimoto, Hisaichi Shibata, Hiroshi Koizumi, Yuichi Kuya, Tomohide Inari, Ryota Hirashima, Yoshiharu Tamaki, Takuya Karatsu

Abstract

We develop a high-speed/high-precision computational program using a quasi-first principle method, which can faithfully reproduce the actual flight environment to understand the true nature of fluid phenomena. Specifically, we develop a high-precision compressible flow solver with geometric wall models and LES (Large Eddy Simulation) wall models based on hierarchical, orthogonal and equally spaced structured grids.

Reference URL

Please refer to ‘サブ課題D｜重点課題８：近未来型ものづくりを先導する革新的設計・製造プロセスの開発‘.

Reasons and benefits of using JAXA Supercomputer System

We need large computer like JSS2 because our calculations must be large scale computations. Moreover, JSS2 has a similar architecture to the our target computer FUGAKU.

Achievements of the Year

We proceeded with the development of a compressible fluid analysis program FFVHC – ACE using a hierarchical, orthogonal and equally spaced structured grid method.

In this fiscal year, larger scale analysis with approximately up to 4.5 billion grid points, was performed, in comparison with the trial calculation of the detailed geometry of the actual aircraft (JSM_CRM_LEG model) conducted last year,confirming that large-scale analysis was possible. Figure 1 shows the calculation results for a Mach number of 0.2, a Reynolds number of 10 6, and an angle of attack of 7 degrees. In this figure, the vortex around the JSM_CRM_LEG model is shown. Figure 2 compares the results of the last year’s survey with about 800 million grid points and the results of this year’s about 4.5 billion grid points. It can be seen that finer vortices are captured as the grid resolution increases.

Fig.1(video): Flow around detailed geometry model(JSM_CRM_LEG model)

Fig.2: Fine vortex reproducibility due to differences in grid resolution(800 million grid points and 4.5 billion grid points)

Publications

– Non peer-reviewed papers

R. Takaki, S. Kawai, Y. Fukushima, Y. Tamaki, S. Tsutsumi and H. Shibata, Development of a high-speed and high-precision turbulent flow solver

using hierarchical cartesian grids, pp165-171, JAXA-SP-19-007

– Invited Presentations

High-speed tuning of fast flow analysis program – from FX100 to FUGAKU -, 132nd Computational science colloquium

– Oral Presentations

1) R. Takaki, S. Kawai, Y. Fukushima, Y. Tamaki, S. Tsutsumi and H. Shibata, Development of a high-speed and high-precision turbulent flow solver

using hierarchical cartesian grids, 51st Fluid Dynamics Conference/37th Aerospace Numerical Simulation Symposium, 2A01

2) R.Takaki, High-speed and high-precision turbulent flow solver using hierarchical cartesian grid: FFVHC-ACE, 1st Application Collaboration Development Conference

3) R.Takaki, Performance evaluation and high-speed tuning of FFVHC-ACE, 2st Application Collaboration Development Conference

4) R. Takaki, High-speed tuning of FFVHC-ACE, 3rd Integrated Workshop for Manufacturing with HPC

Usage of JSS2

Computational Information

• Process Parallelization Methods: MPI
• Thread Parallelization Methods: OpenMP
• Number of Processes: 1 – 2048
• Elapsed Time per Case: 300 Hour(s)

Resources Used

Fraction of Usage in Total Resources*1(%): 4.52

Details

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

*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 2019-March 2020