Japan Aerospace Exploration Agency Chooses HPC From Fujitsu

Fujitsu Ltd. received an order for a HPC system from the Japan Aerospace Exploration Agency (JAXA), as the organization plans to expand its use in the HPC field in addition to conventional numerical simulations.

Accurately reproduces the actual flight environment
and calculates the vortices in the space that are
generated around the aircraft ^(*7)

Scheduled to start operation in October 2020, the new computing system for large-scale numerical simulation, composed of Fujitsu PrimeHPC FX1000 ⁽¹⁾, is expected to have a theoretical computational performance of 19.4 petaflops ⁽²⁾, which is approximately 5.5 times that of the current system. At the same time, Fujitsu will implement 465 nodes of x86 servers the company’s server Primergy series for general-purpose systems that can handle diverse computing needs.

Calculates the generation and propagation mechanisms
of engine exhaust jets and acoustic waves during rocket launch

The system will contribute in improving the international competitiveness of aerospace research, as it will be used as the basis for JAXA’s HPC. It is also expected to be used for various applications, including a large-scale data analysis platform for satellite observation and an AI calculation processing platform for joint research.

Developmental background
As it conducts research of space development, aviation technology, and related basic technology, JAXA has used HPC systems to develop numerical simulation technologies such as fluid dynamics and structural dynamics in the study of aircraft and rockets. In recent years, in addition to conventional numerical simulations, the system has been expanding their role in the HPC field. For example, the system has processed earth observation data collected by satellites for use by researchers and the general public, while it has been used in AI calculations, including deep learning.

JAXA is currently operating a HPC system JSS2 ⁽³⁾ comprised of SORA-MA ⁽⁴⁾, which consists 3,240 nodes of Fujitsu HPC PrimeHPC FX100, and J-SPACE ⁽⁵⁾ that stores and manages various data using a large-capacity storage medium.

PrimeHPC FX1000 (1 rack: 384 nodes)

Features of HPC system
As a successor to the JSS2 system, Fujitsu will implement a computing system for large-scale numerical simulations. The system will consist 5,760 nodes of PrimeHPC FX1000, which utilizes the technology of HPC Fugaku jointly developed by Fujitsu and Riken. It is expected to have 19.4 petaflops, approximately 5.5 times the theoretical computing performance of the current system, in double precision (64bit) usually used in simulations. In addition, a total of 465 nodes from x86 servers Primergy series equipped with high memory capacity and GPU ⁽⁶⁾ will be deployed as they compose a general-purpose system capable of handling a variety of computing needs. With a large file system capacity of approximately 50PB, including high-speed access storage system of approximately 10PB, the new system will offer high performance and ease of use.

The implementation of PrimeHPC FX1000 equipped with a n ARM architecture CPU A64FX will enable the application of various software and contribute to the widespread use of JAXA’s research results.

Future plans
While enhancing the global advantage of JAXA’s aerospace research in the conventional numerical simulation field, the system, as the foundation of the Agency’s HPC infrastructure, will be used for an AI computational processing platform for joint research and shared use. The system will also be applied to a large-scale data analysis platform for aggregating and analyzing satellite observation data that had been previously stored and managed by different divisions at JAXA.

Fujitsu will support JAXA in making its philosophy a reality by solving its issues with experience gained through supplying HPC systems to the Agency since the 1970s. Offering PrimeHPC FX1000 worldwide, the company will contribute in solving social issues, accelerating research, and bolstering the competitive edge of corporations.

Main configuration of system:

Computing system
Click to enlarge

General-purpose system
Click to enlarge

Parallel file system

Comment by Naoyuki Fujita, manager, HPC division security and information systems departmentt: “JAXA: A large-scale numerical simulation environment has been essential in JAXA’s long ranging R&D in the aerospace field, and the currently operating JSS2 has greatly contributed to such progress. We expect that the new system will further improve simulation performance with PrimeHPC FX1000, which utilizes Fugaku technology that excels in large-scale research. Moreover, we hope that it will play an active role in research in new fields such as AI.”

Resources:
Japan Aerospace Exploration Agency (JAXA)
About JSS2      
Fujitsu HPC PrimeHPC          
Fujitsu Server Primergy RX2540 M5        
Fujitsu Server Primergy CX2570 M5

^[1] Fujitsu HPC PrimeHPC FX1000
Features the CPU A64FX adopting Scalable Vector Extension (SVE) which is an extension of the Armv8-A architecture for HPCs. It achieves high performance per power and scalability.
^[2] Petaflops
Short for Peta Floating-point Operations Per Second. It is a unit of computing power that indicates the number of floating-point operations per second. Peta stands for 1,000,000,000,000,000 (15 raised to the power of 10).
^[3] JSS2
Short for JAXA HPC System gen 2. This second-gen HPC system at JAXA has been in operation since April 2016.
^[4] SORA-MA
Short for HPC for earth Observation, Rockets, and Aeronautics (Main). It is the main system in JSS2 and has a theoretical computational performance of 3.49 petaflops. Highly tuned program assets can be executed in highly parallel environments and on a large scale.
^[5] J-SPACE
Short for Jaxa’s Storage Platform for Archiving, Computing, and Exploring. Various data can be archived, as High Performance Storage System (HPSS) manages 3PB disk and 60PB tape media in tiers.
^[6] GPU
Short for Graphic Processing Unit, which applies hardware used for image processing to both image processing and numerical calculation. With high computing performance, GPU is increasingly being used in HPCs.
^[7] Image simulated with the current system ‘JSS2’ (flight environment)
As part of the ‘Social and scientific priority issues to be tackled by using the HPC Fugaku’ promoted by the Ministry of Education, Culture, Sports, Science and Technology (MEXT), this initiative represents the Sub-issue D ‘R&D of core technologies to innovate aircraft design and operation’ under the Priority Issue 8 ‘Development of innovative design and production processes that lead the way for the manufacturing industry in the near future.’
^[8] Node configuration
XL node, L node: Node for non-parallelized specialized application. There are two types, XL and L, depending on the capacity of the main memory. M node: Node that perform computational processing for such specialties of GPUs as AI and Deep Learning. S node: Node that runs an application in high parallel. This is for general applications such as commercial applications.