ILLINOISROCSTAR LLC — Department of Energy SBIR Phase I: 09a

Description

Creation of national web–enabled infrastructure for predictive theory and modeling is needed to facilitate the coordination and sharing of information and data, scalable codes, and for their implementation on or transfer to new architectures. In addition, a web–based infrastructure is needed to impose universal standards for data inputs and outputs in the multitude of codes and methodologies or to capitalize upon semantic strategies for bypassing the need for universal standards altogether. Coordination, standardization, and sharing of chemistry and materials data are being addressed in this project through construction of a distributed web-based infrastructure specifically designed to provide standardized storage, organization, and retrieval of chemical and materials predictive simulation datasets and resultsets. Standardization is being addressed through use of a chemistry and materials–specific markup language that will be used within the site, and promoted as a data exchange mechanism external to the site. File-level storage and metadata collection are provided for breadth of application, while research into tool and area-specific dataset and result parsing will provide depth of coverage. The ability to archive and protect sensitive or proprietary data is also needed to produce the desired national use in order to collect and make available the largest set of open data to promote U.S. competitiveness. Commercial Applications and Other Benefits: The goal of assembling and making available an open, nationally-published repository of chemistry and materials datasets and results will drive a number of commercial applications, as well as national benefits. Ultimately, the distributed system will consist of a main open repository for public data, and a satellite ecosystem of private repositories storing private data, but interacting with the public instance. In this way, industry, government, and academia will be able to take advantage of the open data, contributing as they are able, while still protecting sensitive or proprietary data. This is the only way that a truly national web of connected data will be able to be generated. Once this organization of data begins to grow, the benefits of decreased duplication, increased interoperability, increased speed of accessing needed data, and ultimately decreased cost and time to market for U.S. industrial applications will be realized. Duplication of effort in discovering data for predictive simulation wastes resources nationally that could be better spent speeding delivery of new material and molecular results for clean energy. A nationally accessible web of chemistry and materials data repositories will facilitate inter-organizational sharing, accelerating breakthroughs across industry, government, and academia.