Near-Real Time Networked Analysis of KSTAR Data
July 7, 2020
Scientists and engineers have established a near real-time networked analysis of data taken at the KSTAR fusion experiment in South Korea. The framework is using Machine Learning and Artificial Intelligence algorithms running at NERSC for analysis of large data streams.
Significance and Impact
Fusion reactors have the potential to provide a huge source of clean and affordable electricity. Experiments are underway around the world to turn this promise into reality. These new capabilities allow United States fusion researchers to have broader and faster access to KSTAR (and future ITER) data, enabling faster analysis and informed steering of the fusion experiments.
- The end-to-end Python framework DELTA streams data using ADIOS DataMan over WAN (at rates > 4 Gbps), asynchronously processes on multiple workers with MPI & multi-threading
- Implemented deep convolutional neural networks for working with multi-scale fusion data, e.g. ECEi, for recognizing events of interest.
- With KSTAR data streaming to NERSC, the time for an ECEi analysis was reduced from 12 hours on single-process to 10 minutes on 6 Cori nodes.
R. Kube, RM Churchill, J. Choi, et al., SciPY 2020 Conference Proceedings, “Leading magnetic fusion energy science into the big-and-fast data lane,” https://doi.org/10.25080/Majora-342d178e-013 .
About NERSC and Berkeley Lab
The National Energy Research Scientific Computing Center (NERSC) is a U.S. Department of Energy Office of Science User Facility that serves as the primary high-performance computing center for scientific research sponsored by the Office of Science. Located at Lawrence Berkeley National Laboratory, the NERSC Center serves more than 7,000 scientists at national laboratories and universities researching a wide range of problems in combustion, climate modeling, fusion energy, materials science, physics, chemistry, computational biology, and other disciplines. Berkeley Lab is a DOE national laboratory located in Berkeley, California. It conducts unclassified scientific research and is managed by the University of California for the U.S. Department of Energy. »Learn more about computing sciences at Berkeley Lab.