In EDS ansehen

Very large-scale diffraction investigations enabled by a matrix-multiplication facilitated radial and azimuthal integration algorithm: MatFRAIA

Gespeichert in:
Bibliographische Detailangaben
Titel: Very large-scale diffraction investigations enabled by a matrix-multiplication facilitated radial and azimuthal integration algorithm: MatFRAIA
Autoren: Jensen, Alexander Bernthz, Christensen, Thorbjørn Erik Køppen, Weninger, Clemens, Birkedal, Henrik
Weitere Verfasser: Lund University, Profile areas and other strong research environments, Strategic research areas (SRA), eSSENCE: The e-Science Collaboration, Lunds universitet, Profilområden och andra starka forskningsmiljöer, Strategiska forskningsområden (SFO), eSSENCE: The e-Science Collaboration, Originator, Lund University, MAX IV Laboratory, Lunds universitet, MAX IV-laboratoriet, Originator
Quelle: Journal of Synchrotron Radiation. 29:1420-1428
Schlagwörter: Natural Sciences, Computer and Information Sciences, Computer Engineering, Naturvetenskap, Data- och informationsvetenskap (Datateknik), Datorteknik
Beschreibung: As synchrotron facilities continue to generate increasingly brilliant X-rays and detector speeds increase, swift data reduction from the collected area detector images to more workable 1D diffractograms becomes of increasing importance. This work reports an integration algorithm that can integrate diffractograms in real time on modern laptops and can reach 10 kHz integration speeds on modern workstations using an efficient pixel-splitting and parallelization scheme. This algorithm is limited not by the computation of the integration itself but is rather bottlenecked by the speed of the data transfer to the processor, the data decompression and/or the saving of results. The algorithm and its implementation is described while the performance is investigated on 2D scanning X-ray diffraction/fluorescence data collected at the interface between an implant and forming bone.
Zugangs-URL: https://doi.org/10.1107/S1600577522008232
Datenbank: SwePub
Beschreibung
Abstract:As synchrotron facilities continue to generate increasingly brilliant X-rays and detector speeds increase, swift data reduction from the collected area detector images to more workable 1D diffractograms becomes of increasing importance. This work reports an integration algorithm that can integrate diffractograms in real time on modern laptops and can reach 10 kHz integration speeds on modern workstations using an efficient pixel-splitting and parallelization scheme. This algorithm is limited not by the computation of the integration itself but is rather bottlenecked by the speed of the data transfer to the processor, the data decompression and/or the saving of results. The algorithm and its implementation is described while the performance is investigated on 2D scanning X-ray diffraction/fluorescence data collected at the interface between an implant and forming bone.
ISSN:16005775
DOI:10.1107/S1600577522008232