XIMG: A Materials Science Gateway for X-ray imaging and modeling of microstructures

We describe the development and deployment of a Science Gateway that includes a set of tools, computational materials-science models, and x-ray scattering data that are integrated and available through a single web-based location. It is well recognized that advances in the development of new structural materials with specified properties require improved diagnostics across all the relevant length and time scales. Examples of technological needs include: development of high strength, high ductility, low density metals for light weight applications in a vast variety of aerospace, power generation, and structural materials; accurate predictions of residual stress and relaxation and recovery in additive manufacturing and metal forming, for example in the automobile and defense industries; and improved reliability of interconnects in semiconductor chip technology. Advanced x-ray diffraction imaging at leading synchrotron facilities uniquely provides the necessary diagnostic information about the three-dimensional structure of complex materials with unprecedented resolution. Key aspects of the technique include being non-destructive in hard materials, and fast enough and of sufficient resolution to observe structural responses at the sub-micron scale in samples as they are subjected to thermal, mechanical, or other types of loading. Such unprecedented resolution comes with very large datasets, uniquely rich in information, but difficult to mine and analyze, except by a very small number of experts. This is a significant bottleneck to progress.

This program develops an open cyberinfrastructure in the form of a public Science Gateway to serve the Cornell High Energy Synchrotron Source. The gateway is based on the Galaxy framework. Raw data is accumulated locally at the beam lines, and ingested into a Galaxy instance using standard web application programming interfaces. Data is typed, metadata is attached, and is made available through a shared data library. The gateway then provides the infrastructure for user defined transformations, including data reduction, image reconstruction, and feature analysis, while retaining metadata and provenance information. The gateway supports data transfer to XSEDE resources, as Galaxy natively orders individual codes according to category, and maps the execution of the designed workflows to the necessary resources without user awareness of their location. Workflows including visualization and modeling are also supported. The gateway is extensible by users of the facility and the broader community as new analysis and modeling tools are developed.