Optimizing inference of segmentation on high-resolution images in MLExchange
The Journal of Supercomputing · 20.06.2025 · 81 (9)
DOI: 10.1007/s11227-025-07413-5 ↗Kurz gefasst
Fachartikel in der Zeitschrift The Journal of Supercomputing, online erschienen am 20.06.2025. Verfasst von Shizhao Lu und 5 weiteren Autor:innen. Beteiligte Einrichtung: Advanced Light Source, Lawrence Berkeley National Laboratory, Berkeley, CA 94720 USA. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 29 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NIGMS NIH HHS, Basic Energy Sciences, U.S. Department of Energy, Scientific User Facilities, U.S. Department of Energy.
Abstract (Originalsprache)
MLExchange is a machine learning (ML) operations platform providing web user-interfaces (UIs) for data visualization and analysis pipelines at synchrotron facilities. Among these UIs is the segmentation app which helps synchrotron users utilize ML algorithms to automatically segment high-resolution scientific images with minimal manual annotation effort. In this work, we share code optimizations that significantly speed up the segmentation inference workflow of large data in short time. By optimizing the sequence of CPU-GPU data transfers and introducing CPU parallelization to key operations, we improve the per-device, per-image frame computational efficiency and observe close to 3 $$\times$$ × speedup over the original segmentation inference workflow run time when utilizing a single GPU. Further adaptations enabling multi-GPU inference yield more than 40 $$\times$$ × speedup with 100 GPUs compared to the optimized single GPU inference workflow. This acceleration of the segmentation inference workflow will provide MLExchange users with easy access to segmentation results with little wait time.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Advanced Light Source, Lawrence Berkeley National Laboratory, Berkeley, CA 94720 USA.
- Center for Advanced Mathematics for Energy Research Applications, Lawrence Berkeley National Laboratory, Berkeley, CA 94720 USA.
- Molecular Biophysics and Integrated Bioimaging Division, Lawrence Berkeley National Laboratory, Berkeley, CA 94720 USA.
- Berkeley Synchrotron Infrared Structural Biology Program, Lawrence Berkeley National Laboratory, Berkeley, CA 94720 USA.
Förderung
- NIGMS NIH HHS
- Basic Energy Sciences, U.S. Department of Energy
- Scientific User Facilities, U.S. Department of Energy
- National Institutes of Health
Zitieren
APA
Lu, S., Chavez, T., Koepp, W., Hao, G., Zwart, P. H., & Hexemer, A. (2025). Optimizing inference of segmentation on high-resolution images in MLExchange. The Journal of Supercomputing, 81(9). https://doi.org/10.1007/s11227-025-07413-5
Vancouver
Lu S, Chavez T, Koepp W, Hao G, Zwart PH, Hexemer A. Optimizing inference of segmentation on high-resolution images in MLExchange. The Journal of Supercomputing. 2025;81(9). doi:10.1007/s11227-025-07413-5
BibTeX
@article{Lu2025,
author = {Lu, Shizhao and Chavez, Tanny and Koepp, Wiebke and Hao, Guanhua and Zwart, Petrus H and Hexemer, Alexander},
title = {Optimizing inference of segmentation on high-resolution images in MLExchange},
journal = {The Journal of Supercomputing},
year = {2025},
volume = {81},
number = {9},
doi = {10.1007/s11227-025-07413-5}
}
EUVIMED