[User's Voice] National Institute for Fusion Science

PDF Download

Create a sun on Earth!

There is growing expectation that nuclear fusion will be a clean source of energy in the future. Research is actively promoted in this field and experiments are performed at the cutting edge of nuclear fusion technology. The National Institute for Fusion Science (NIFS) generates more than 5 TB of data every day. The Panasonic Data Archiver stores this huge amount of data on the same day and ensures that the valuable information is accessible by researchers around the world.

Located in Japan, NIFS is one of the world's leading research institutes in this scientific field. Using the world's largest-class experimental machine, a Large Helical Device (LHD), NIFS conducts experiments to create ultra-high-temperature plasma higher than 100 million degrees Celsius for achieving nuclear fusion. At NIFS, as many as 180 experiments are conducted per day, and the amount of data obtained in a day exceeds 5 TB. The Panasonic Data Archiver stores this huge amount of data on a daily basis. Why does NIFS need to archive such a large volume of data? To find out, we interviewed Associate Professor Hideya Nakanishi and Mr. Masaki Osuna, who have been engaged in research and data management for many years at NIFS.

Dr. Hideya Nakanishi
(Ph.D. in Engineering)

Associate Professor in the High-Temperature Plasma Physics Research Division and Task Leader in the Division of Information and Communication Systems
National Institute for Fusion Science

Mr. Masaki Osuna

Diagnostic Technology Division, Department of Engineering and Technical Services
National Institute for Fusion Science

Attempting to create a new, dreamed up energy source from ultra-high-temperature plasma exceeding 100 million degrees Celsius

Nuclear fusion is a reaction in which the nuclei of lightweight atoms, such as hydrogen, fuse and create heavy atoms. In this process, an immense amount of energy is discharged. The sun continues to burn because of the nuclear fusion that takes place at its core. Achieving nuclear fusion would enable us to "create a sun" on Earth. If we succeed, we can make a giant leap toward the acquisition of a dreamed up energy source, that will provides us with a virtually unlimited amount of energy. The fuels necessary for nuclear fusion are hydrogen and lithium, both of which are contained abundantly in seawater. By using a small amount of seawater, nuclear fusion can generate and supply enough electricity to meet the entire power demand of Japan throughout the year. Thus, this new energy source is expected to replace the fossil fuels we now rely on.

NIFS, which is located in Toki City, Gifu Prefecture, is equipped with the world's largest-class superconducting stellarator, Large Helical Device (LHD). It is one of the world's leading research institutes in this scientific field. The LHD is used to generate ultra-high-temperature plasma exceeding 100 million degrees Celsius that is necessary for achieving nuclear fusion and to collect various fundamental data for use in research and development activities. Backed by many years of experience in conducting experiments, Dr. Nakanishi describes the features of the LHD.

"The LHD generates a powerful spiral-shaped magnetic field and confines ultra-high-temperature plasma. Since the machine uses superconducting magnets cooled to a temperature of -269℃ to generate the magnetic field, it does not produce heat. This enables us to repeat the same experiment in short intervals of once every three minutes. If we did not use the superconducting magnets, a large amount of heat would be produced every time the electric current flowed, thus requiring time to cool before conducting the next experiment."

There is a plan to start experiments in fiscal 2016 to generate plasma using deuterium (heavy hydrogen) instead of light hydrogen that enables nuclear fusion easier than conventionally. This will take the experiments at NIFS to the next stage.

* Plasma confinement methods can be categorized into two types: helical and tokamak, depending on the shape formed by the installed magnets. The machine at NIFS is a helical type with spirally installed magnets, which was developed originally in Japan. Compared to the tokamak type, the helical system can operate stably for a long time, so it is said to be suitable for use in commercial furnaces in the future.

Archiving huge amounts of data and ensuring uninterrupted service to provide them to researchers the world over

Each experiment conducted with the LHD results in an immense amount of data, totaling about 30 GB. Since as many as 180 experiments are conducted in a day, a maximum of 5.3 TB of data is created per day. The data output from more than 100 measuring instruments mounted on the machine is compressed by the data collecting computers, then written onto hard disks as well as optical discs in the Panasonic Data Archiver. Tape could not be used as a data recording media for archiving purposes for a number of reasons, according to Dr. Nakanishi.

"We provide all accumulated experiment data on-line to researchers around the world. For this service, our main storage system uses hard disks. If a hard disk problem occurs, it is necessary to switch to the data archiving system to provide data from there. The most important thing is to deliver data without any interruptions. Therefore, the data archiving system must use random access media. If tape were used, it would have to be replaced about every three years. That would increase the running cost. In my experience, I wouldn't consider tape as long-term data storage media. For long-sustained research, data must be stored much longer than several decades, and presently there is no other choice but optical discs." (Dr. Nakanishi)

History of struggles with the management of colossal data behind the advancement of research

The history of NIFS is also the history of struggles with the enormous amount of data generated every day. Mr. Masaki Osuna in the Diagnostic Technology Division, Department of Engineering and Technical Services recalls the difficulties he faced.

"In the early years, we were using 4.8-GB MO discs for archiving data. In about 2001, we began using less expensive DVD. However, DVD generated more errors than we had expected, so it presented a reliability issue. Furthermore, the amount of output data increased year after year, so the number of DVDs used in a week exceeded 100 and writing one day's data onto DVDs took more than 24 hours, which was approaching the time limit. Then, for about two years until the release of the Blu-ray Disc, we used UDO media (Super MO discs) to barely make it through. In 2006, we started using 50-GB Blu-ray discs. Presently, we operate three units of Panasonic Data Archivers and store data on 100-GB discs."

NIFS chose the Panasonic Data Archiver because of its large capacity, high reliability, long life and random access capability. Now the Data Archiver plays an important role at NIFS. However, the amount of data generated at NIFS continues to increase to no end. Dr. Nakanishi said that the amount of generated data was growing by one digit every five years.

Researchers' insatiable curiosity and inquisitiveness results in explosive data growth

Why is it necessary to collect and archive such an enormous amount of data in the first place? Dr. Nakanishi gives an answer to this question.

"If the amount of acquired data increases tenfold, a phenomenon that was previously perceivable only faintly through the mist becomes clearly visible. Behavior of plasma that we did not previously know can be perfectly understood. To that end, we consistently improve the performance of the measuring instruments mounted on the machine. What drives us to do so is none other than the insatiable curiosity and inquisitiveness of researchers."

NIFS has been archiving all experiment data ever since it conducted its first experiments in 1998. Dr. Nakanishi explains that the fact that all data is archived is of great significance.

"Experiments for plasma generation are affected by many factors, including the conditions of the equipment and devices, so it is difficult to perfectly reproduce an experiment conducted in the past. For example, even a slight change in the condition of the inside surface of the vacuum container used for plasma generation, such as the adhesion of impurities, can affect plasma generation and produce a completely different result in some cases. Therefore, each and every piece of past data is valuable. Furthermore, even if the data of one experiment does not provide the clear picture of a phenomenon, statistical analysis of data obtained from many experiments can shed light on the phenomenon in many cases. Therefore, possessing many data parameters and being able to examine a large volume of past data are critical."

Growing demand for an even larger storage capacity created by next-generation nuclear fusion experimental devices

The amount of data generated every day at NIFS is growing at an accelerated rate, and the Panasonic Data Archiver continues to archive it reliably. To respond to future advancement, the Data Archiver needs to evolve to the next stage.

"The next-generation International Thermonuclear Experimental Reactor (ITER) project is now underway in France. It is an international nuclear fusion research and engineering megaproject undertaken jointly by Japan, EU, United States, Russia, China, Korea and India. The facility is scheduled to begin operation from around 2020, and the project costs nearly two trillion yen in total. The experimental machine is larger than the one at NIFS, and the generated data will also be greater in volume. Archiving that data requires storage media with a larger capacity for sure."

There is a plan to develop higher densities of 500-GB and one-terabyte Archival Discs in the near future. To meet the high-level demands of ITER, it is necessary to further advance the technologies at an accelerated pace not only for the data recording media but also for the system.

Nuclear fusion experiments are a proving ground for achieving the grand human dream of possessing an inexhaustible source of energy. To make this dream come true, numerous researchers are conducting experiments day after day and grappling with huge amounts of data output from experiments. Panasonic continues to help make the dream come true by making technological innovations for optical disc-based data archiving systems.

User Information:

National Institute for Fusion Science
322-6, Oroshi-cho, Toki-shi, Gifu 509-5292, Japan

* This article is based on an interview held in May 2016.

Related Articles:

Page top