19^th International CODATA Conference
Category: Poster, Materials

Virtual Experiments Platform for Materials Design and Modeling

Nobutaka Nishikawa (nisikawa@ccse.fuji-ric.co.jp), Materials Science and Nanotechnology, Science and Technology, Fuji Research Institute Corporation, Japan
Masatoshi Nihei (NIHEI.Masatoshi@nims.go.jp), Materials Engineering Laboratory, National Institute for Materials Science, Japan
Shuichi Iwata (iwata@k.u-tokyo.ac.jp), Graduate School of Frontier Sciences, The University of Tokyo, Japan

Current status of our virtual experiment platform is reported as one of progress reports on ITBL (Information Technology Based Laboratory)  Project. On this platform, various components of computational materials science and engineering technologies are synthetically integrated, such as simulation, database, data analysis, visualization, and human interface. A special feature of this platform is the realization of a new concept by using the "taskflow engine" to control and to operate various components of computational technology.

In this taskflow engine, all computational modules for materials design are automatically linked for materials design along a predefined task scenario.    Recent progresses in high performance computing and networking give us  huge possibilities for ad hoc integration of data and models, and sometimes for designing materials as a consequence of massive trial and errors. Many components are expected to be developed for materials design such as simulation models, databases, knowledge management, visualization and the GUI.

One of the difficulties along the taskflow is the fact that data and models are not prepared to form a continuous flow along each task in general. In this paper, we would like to show this difficulty by using such examples as superalloys, magnetic materials, which are followed by explanations on ad hoc tentative solutions to drive each design procedure.