Reconstruction: Move selected instrument data to HPC for full first-principles reconstruction.Indexing: Extract and synthesize metadata, and load into catalog.Analysis: Process ML model results to derive additional quantities.Data capture: Retrieve data from instrument, perform fast reconstruction with edge ML model, and stage results to temporary storage.The Braid project aims to overcome these challenges by making it easy for researchers to define sets of flows that individually and collectively implement application capabilities while satisfying requirements for rapid response, high reconstruction fidelity, data enhancement, data preservation, model training, etc.įor example, a ptychography experiment that uses a machine learning (ML) model for rapid reconstruction may be structured as five distinct flows: The result is missed scientific opportunities, both during experiments and later due to data not being rendered findable, accessible, interoperable, and reusable (FAIR). ![]() However, the volume, velocity, and variety of data produced by these instruments challenge today’s often manual data collection, analysis, and curation methods. Next-generation scientific instruments offer new means to understand and harness a broad range of physical and biological phenomena. ![]() ![]() Braid Braid: Data Flow Automation for Scalable and FAIR Science
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |