【Success Story Vol. 46】
Kento Tanaka, Assistant Professor at Okayama University

■ Learning Experience

In March 2021, Kento Tanaka graduated from the Graduate School of Engineering at Nagoya University, specializing in aerospace engineering. In April of the same year, Kento Tanaka started his position as an Assistant Professor in the Faculty of Engineering at Okayama University. He conducts research in the fluid mechanics laboratory, focusing on the effects between shockwaves and turbulence, and also provides guidance to students.

■ Main Research

His expertise is in aerospace engineering, specifically focusing on the "effects of shockwaves on turbulence" in fluid mechanics. When an aircraft flies at supersonic speeds, the fluid in front of it gets compressed, generating shockwaves. When these shockwaves reach the ground, they produce explosive sounds similar to fireworks, leading to noise-related issues. To elucidate this mechanism, he is studying how shockwaves generated in the upper atmosphere can interfere with the characteristics of atmospheric flow as they propagate.

There are two research methods involved: reproducing atmospheric flow by generating wind physically and simulating atmospheric flow on a computer. This is Kento Tanaka’s area of expertise. First, he investigates the locations and types of noise occurrences. In the future, his goal is to provide recommendations for the prevention of aircraft noise and the development and operation of supersonic passenger aircraft.

■ Why He Needs a NAS

After coming to Okayama University, Kento Tanaka started using neural networks to analyze data, and there are also students in my laboratory conducting similar research. He aims to train numerical simulation data and eventually apply it to experimental data. However, the fundamental requirement is that he will be dealing with a large amount of data and will need ample storage space to store it. This aligns with the growing trend in data-intensive research, especially when utilizing neural networks for analysis.

Up until now, when data was transferred from students conducting numerical computations to students performing neural network analysis, data exchange was done using HDDs. This process consumed a significant amount of time and effort and required frequent HDD purchases. Therefore, Kento hopes to streamline data sharing among other students by enabling them to download and store data on the NAS. Even if there are data-related issues, they can be easily addressed by checking the NAS, eliminating the hassle of plugging and unplugging hard drives. The reason it was decided to introduce it was to facilitate smoother data sharing within the laboratory. The high storage capacity enables seamless and scalable data sharing, essentially allowing you to centralize data for the entire laboratory through a single NAS.

■ The Solution

“Since the NAS catalog is regularly delivered to the laboratory, I discussed it with other professors and decided to introduce the NAS as described in the catalog. It was a somewhat impromptu decision, but I'm pleased that they provided me with detailed explanations. If possible, I would appreciate it if we could skip the initialization setup process, as it would be a great help if the product comes with the initialization already completed. While the operation is ongoing, I cannot focus on my research if the computer acting as a tool experiences issues. Therefore, it would be very helpful if we could start using it immediately.”

-Kento Tanaki

■ How He Uses His NAS

Absolutely, for analyses involving neural networks and other data-intensive tasks, having the capacity to add dedicated machines can significantly enhance your ability to process more data efficiently. Simply storing data without the means to analyze it would indeed be wasteful. Therefore, expanding the number of machines capable of data analysis is a logical and beneficial step in your research process.

Enabling students to easily access each other's data while they work can indeed help prevent accidental data deletion, provided that every student understands how to use it. In fact, having open data can promote progress in new analyses, which can be a positive aspect of data science. Sharing data and allowing collaboration can lead to innovative research and foster a culture of knowledge sharing and learning. It's a good practice for a collaborative and data-driven environment like a research lab.

As data science and data analysis continue to advance, the results of research in this field are increasingly applied in our daily lives. When considered, a NAS is indeed a valuable tool for researchers as it allows for the storage of large volumes of data and facilitates real-time data sharing. This can greatly enhance collaboration, research efficiency, and the overall progress of scientific endeavors.


ソース: https://bto.applied.ne.jp/hpc_case_study_46.html