Presenter Profile

Education:

Position

Research Project and Interests:

Awards:

KEYNOTE SPEAKER
Roger H. French

Professor
Case Western Reserve University, Department of Materials Science and Engineering

Roger H. French is the Kyocera Professor in the Case School of Engineering, Case Western Reserve University, Cleveland, Ohio. His primary appointment is in Materials Science and Engineering, with secondary appointments in Computer and Data Sciences,Macromolecular Science, Biomedical Engineering, and Physics. He is the director of the DOE-NNSA Center of Excellence for Materials Data Science for Stockpile Stewardship. He is co-PI of the Center for Advancing and Distributed Fertilizer Production (CASFER), an NSF sponsored IUCRC. He is the faculty director of the CWRU Applied Data Science program which offers graduate courses and graduate certificates and an undergraduate minor university-wide. 

TALK TITLE
Accelerating Time to Science using CRADLE, Cyberinfrastructure for Materials Data Science

KEYWORDS
Materials Data Science, Distributed Computing, geospatiotemporal, SXRD

ABSTRACT
Modern materials science research can generate Petabyte-scale spatiotemporal datasets that span a number of modalities and formats, presenting a challenge to data science and analytics. Creating an accessible computing infrastructure and frameworks that supports the scale and diversity of materials science data for scientists to use is a non-trivial task. 

We have developed the Common Research Analytics and Data Lifecycle Environment (CRADLE) to address the challenges of materials data science through a scalable research computing framework and cyberinfrastructure. CRADLE integrates Petabyte (scaled-out) distributed computing with Petaflop (scaled-up) high performance computing to enable materials scientists to coherently learn from sparse to massive datasets, using GPU training of AI and deep learning models for complex multimodal datasets and problems. CRADLE can 1) utilize a myriad of low to high performance computer systems; 2) be accessible to research scientists with limited to extensive computational backgrounds; 3) ingest and curate large-scale, heterogeneous datasets and their resulting deep learning models and AI agents  4) provide a flexible toolbox for building automated AI / machine learning pipelines that span from ingestion through exploratory data analysis, model training to deployment. 

This coherent materials data science cyberinfrastructure and data lifecycle environment is exemplified by FAIRification, graphs learning, geospatiotemporal modeling, digital twins and Terabyte scale analyses of Synchrotron X-ray diffraction (SXRD) and X-ray Computed Tomography (XCT) studies. CRADLE illustrates the path to implementation of AI for Science.