NVIDIA cuPhoton accelerates scientific image analysis

What happened

NVIDIA has released NVIDIA cuPhoton, an open-source CUDA-X toolkit intended to move high-throughput (how much work a system gets through in a given time) scientific image processing away from CPU (the general-purpose processor at the centre of a computer)-bound pipelines and onto GPUs (a chip built for many calculations at once, used for graphics and AI). The software targets workloads including optical astronomy, X-ray analysis and laser experiments, where instruments can generate data far faster than conventional processing chains can analyse it. NVIDIA is a semiconductor company based in Santa Clara, and its products and services include graphics processing units and software.

NVIDIA cuPhoton keeps data on the […] The post NVIDIA cuPhoton accelerates scientific image analysis appeared first on eeNews Europe . NVIDIA cuPhoton keeps data on the GPU The idea is to keep image data resident on the GPU from ingestion through processing, fitting and classification rather than repeatedly transferring it between host and accelerator memory. NVIDIA’s GitHub project repository contains modules for loading FITS and HDF5 data, reprojection, image subtraction, nonlinear fitting, candidate classification and X-ray detector analysis.

Key facts

• NVIDIA has — released: NVIDIA cuPhoton, an open-source CUDA-X toolkit intended to move high-throughput scientific image processing away from CPU-bound pipelines and onto GPUs
• NVIDIA’s GitHub project repository — includes: modules for loading FITS and HDF5 data, reprojection, image subtraction, nonlinear fitting, candidate classification and X-ray detector analysis