Skip to content. | Skip to navigation

Personal tools
Log in
Sections
You are here: Home torrent No.3 The Young Leaders who are Creating the Computational Materials Science of Tomorrow

The Young Leaders who are Creating the Computational Materials Science of Tomorrow

 

CMSI has introduced a new Division Researcher program. Division Researchers have various roles and missions, but the goal of all activities is the promotion and development of computational materials science. In this issue, we focus on how this position came to be created and how the roles of Division Researchers were determined. We will also hear from new researchers who have begun actual activities as Division Researchers to get a sense of their determination and objectives.

 

Turning the Headwaters of Materials Science into a"Torrent"of Innovations

 

The Computational Materials Science Initiative (CMSI) is a research network made up of computational science researchers in the three fields of condensed matter physics, molecular science and materials science. CMSI has received a grant from the Ministry of Education, Culture, Sports, Science and Technology to promote research in Field 2 “New Materials and Energy Creation ” of the High Performance Computing Infrastructure (HPCI) strategy program. The goal of computational materials science is to link theory and experiment to open new doors in science and technology. The computational materials science is a young science, in which new techniques in advanced computing are being developed and whose diverse research topics range from single molecules to practical materials. The aim is to use supercomputers ̶ led by the K Computer, the world’s fastest supercomputer ̶ to expand and deepen the headwaters of materials science into a torrent of innovations in material functions and energy conversion.
What distinguishes the CMSI strategy program is that, in addition to state-of-theart research and development of specific “hot topics,” creating a research and development infrastructure for the next generation of computational materials science is one of the major objectives. To this end, CMSI holds symposiums, workshops, hands-on training sessions, collaborations with experimental researchers and so on to form networks among the people interested in this field, in particular young researchers. CMSI also supports the activities of these researchers whose goal is the formation of a new research and development infrastructure in all areas including computer use, program development, program use and so on. Research achievements as well as personnel, software and simulation data are featured in the CMSI newsletter and on the CMSI website and so on as part of the effort to create a center that can communicate CMSI activities and achievements to the world. Recently CMSI has introduced a unique program, the “Division Researcher” program, for the individuals engaged in activities to promote and develop the field of computational materials science.

 

The Need for a New Research and Development System that Can Respond to Trends in Hardware Development

 

All matter is made up of atoms and molecules. The goal of computational materials science is to determine the behavior of matter as collections of atoms and molecules, and to use computers to solve basic equations that describe this behavior and predict the properties possessed by these materials. Researchers strive to verify experimental results and gain an understanding of the mechanisms by which functions are manifested, and then apply the results in material design.
The computers that are used for simulations are made up primarily of central processing units (CPUs) that perform calculations, memory units that store data and results, and communication units (network) that link the CPUs and the memory units. In order to conduct simulations of physical systems in a larger scale or perform at higher speed calculations that formerly took a great deal of time, supercomputers made up of a network of multiple CPUs and memory units are used. As the pace of acceleration of CPU operating speed has slowed, in recent years the clear trend in large computers has been toward multicore computers and parallel processing. For example, the K Computer has 640,000 computing cores. To make effective use of this type of ultra-parallel computer, revolutionary algorithms and software not achievable through the simple extension of existing technologies must be developed. Until the beginning of the 21st century, increases in computing performance were effected primarily by increasing the operating speed of the individual cores, and it was possible to reap the benefits of improved hardware performance without greatly modifying existing programs. In this sense, it was a time in which one merely had to wait and speeds would increase. But the new trends in computing hardware require a qualitatively different research structure from those of the past. Up to now, software for determining state-of-the-art physicochemical phenomena was almost exclusively developed and maintained on an individual laboratory basis. However, software designed for use on large-scale computers requires a great deal of time and effort for research and development, and improving the program becomes a more complex process. For this reason, software that is useful and can be used by many researchers, and that will contribute to the promotion and development of computational materials science, will need to be maintained and developed organizationally to ensure that the program continues to be passed on to future researchers. It was in an effort to resolve this issue that the “Division Researcher” program was created.

 

Role and Mission of Division Researchers

 

In recruiting and hiring researchers, CMSI makes a clear distinction between “Priority Area Researchers,” who use the K Computer to work on specific key issues in computational materials science, and “Division Researchers,” who work to promote and develop the field of computational materials science itself. Priority Area Researchers endeavor to solve the most urgent problems for which results are needed currently. In contrast, the role of Division Researchers is to find solutions to crucial problems that must be resolved to make progress in next- generation computational materials science, and to pursue infrastructure-building. Their main role is to develop the new methodologies and basic technologies that will be widely used in the field of computational materials science; develop and disseminate important software applications that are expected to be used by many users in the field of computational materials science; develop simulation code by applying the advanced parallelization and speed-up techniques, and conduct computational materials science research using such a code; and build a public release environment to enable many users including both laboratory researchers and company researchers to use research achievements. The goal of all of these missions is to disseminate and expand the academic achievements of stateof-the-art computational materials science research, and ensure that they are returned to society in order to enable computational materials science to be used as a tool for resolving the immediate problems faced by people in their daily lives.
CMSI research topics include explication of the mechanism by which the functions of superconductors and other materials are manifested; generation and storage of next-generation energy, an issue that is expected to become more and more important in the coming years; research to achieve breakthroughs in semiconductor devices, whose rate of speed increase is beginning to plateau; elucidation of molecular control mechanisms for viruses and other threats to humanity; and magnetic materials and structural materials that offer equivalent performance without using scarce elements. All of these areas are ones in which progress is needed to ensure that human beings are able to live lives of abundance here on earth. Computational materials science links theory and experiment in these areas, and Division Researchers in each category (see figure below) conduct research to ensure that achievements in these areas will contribute to academic and social development.

 

拠点研究員e

 

Expanding the Division Researcher Activities Throughout the World

 

The research institutions with which Division Researchers are affiliated include the CMSI Divisions in the areas of condensed matter physics, molecular science and materials science (Institute for Solid State Physics, The University of Tokyo / Institute for Molecular Science, National Institutes of Natural Sciences / Institute for Materials Research, Tohoku University), the Molecular Science Sub-Division (Graduate School of Arts and Sciences, The University of Tokyo), the Materials Science Sub- Division (AIST Kansai, National Institute of Advanced Industrial Science and Technology) and Industry-Government- Academia Cooperation Divisions (National Institute of Advanced Industrial Science and Technology/ National Institute for Materials Science). Moreover, the CMSI Kobe Branch that was established in April 2011 in the Advanced Institute for Computational Science was set up to integrate all of these fields. In addition to Division Researchers, it will have specially appointed CMSI faculty to lead activities to promote individual fields and conduct overall coordination. Division Researcher activities will not be limited to the individual institutions with which they are affiliated. Researchers will be active in all fields of computational materials science and an effort will be made to promote the integration of the condensed matter physics, molecular science and materials science. In addition, technical workshops for young CMSI researchers will be held once every two months. These workshops will feature lectures on the elemental techniques needed for large-scale computing, discussion of the issues faced in each field, exchanges of views regarding the ideal supercomputer in the future and so on. Technical workshops will be planned and conducted by the Division Researchers themselves.
These Division Researcher activities will enhance communication among the young researchers who are the future of computational materials science, with the goal of extending it throughout the entire world.