{"componentChunkName":"component---src-templates-blog-post-js","path":"/blog/morphos-softneuro-enables-19x-faster-inference-of-3d-simulation-on-fugaku/","result":{"data":{"site":{"siteMetadata":{"title":"No Frills News"}},"contentfulNfnPost":{"postTitle":"Morpho’s \"SoftNeuro\" Enables 19x Faster Inference of 3D Simulation on Fugaku","slug":"morphos-softneuro-enables-19x-faster-inference-of-3d-simulation-on-fugaku","createdLocal":"2023-02-09 14:31:04.283043","publishDate":"None","feedName":"Image Recognition","sourceUrl":{"sourceUrl":"https://www.businesswire.com/news/home/20230205005051/en/Morpho%E2%80%99s-SoftNeuro-Enables-19x-Faster-Inference-of-3D-Simulation-on-Fugaku"},"postSummary":{"childMarkdownRemark":{"html":"<p>Comparison of inference speed using TensorFlow (available as standard on Fugaku) and using “SoftNeuro” optimized for Fugaku.\n[Measured Values]Average elapsed time Using \"SoftNeuro\" optimized for \"Fugaku\": approx.\n19.2 times faster (*1) Inference Engine Inference Time (ms) TensorFlow 2820 “SoftNeuro” 147Average power consumption Using \"SoftNeuro\" optimized for \"Fugaku\": approx.\nMorpho will support further acceleration of 3D simulations (galaxy formation simulations) using deep learning on Fugaku through the project and collaboration.\nMorpho will provide broad support, making a wide range of innovations happen with its imaging technology and Deep Learning technology.</p>"}}}},"pageContext":{"slug":"morphos-softneuro-enables-19x-faster-inference-of-3d-simulation-on-fugaku"}}}