Morpho’s ‘SoftNeuro’ Enables 19x Faster Inference of 3D Simulation on Fugaku 

TOKYO, Feb. 9, 2023 -- Morpho, Inc. (Morpho), a global leader in image processing and imaging AI solutions, announced today that it has provided deep learning inference engine “SoftNeuro” to a project promoted by the University of Tokyo, Tohoku University, and Kobe University to accelerate high-resolution galaxy formation simulations using deep learning on the supercomputer Fugaku. The result is approx. 19.2 times faster inference time and approx. 93% reduction in power consumption.

Commenting on the news, Morpho's Masaki Hilaga stated: “The results of this project show that SoftNeuro can significantly improve the calculation speed and reduce power consumption not only on edge devices but also on supercomputers. This indicates that SoftNeuro contributes to the development of science and technology, and also helps to realize the SDGs world through its power-saving performance. Based on these results, we will further promote the technological development of SoftNeuro and expand its range of applications to contribute to the development of our business and society."

Conditions and Measured Values

• “SoftNeuro" is used for 3D-Unet inference on Fugaku.
• Comparison of inference speed using TensorFlow (available as standard on Fugaku) and using “SoftNeuro” optimized for Fugaku.
• Each Fugaku uses 1 node (48 cores).

“SoftNeuro” supports major deep learning frameworks and performs faster processing in various edge-device environments. Since it is a general-purpose inference engine, it can be used not only for image recognition but also for speech recognition and text analysis. Morpho has proposed and provided “SoftNeuro” for multi-platform and high-speed inference for various detection applications based on image data.

Morpho will support further acceleration of 3D simulations (galaxy formation simulations) using deep learning on Fugaku through the project and collaboration. In addition, Morpho will continue to further improve the convenience and technical capabilities of “SoftNeuro” and develop technology on a global level to realize a fruitful culture through the provision of various services and solutions.

About the Project

The project is to accelerate highly resolved galaxy formation simulations. Morpho has developed the 3D-CNN-based deep learning model that predicts anisotropic shell expansion of supernova (SN) explosions and identifies particles with small timesteps. The model is based on Memory-In-Memory Network (Wang et al. 2018), which consists of 2D-CNNs and predicts future images. The research appears in "3D-Spatiotemporal Forecasting the Expansion of Supernova Shells Using Deep Learning toward High-Resolution Galaxy Simulations," Keiya Hirashima, Kana Moriwaki, Michiko S. Fujii (The University of Tokyo), Yutaka Hirai (University of Notre Dame, Tohoku University), Takayuki R. Saitoh, Junichiro Makino (Kobe University).

About Morpho, Inc.

Established in 2004, Morpho (TOKYO: 3653) is a research and development-led company in image processing technology. It has globally expanded its advanced image processing technology as embedded software, for domestic and overseas customers centered on the smartphone market, broadcasting stations and content providers. It has also provided image recognition technology utilizing Artificial Intelligence (AI), collecting image information captured by cameras into devices and clouds and analyzing it, for fields such as automotive devices, factory automation, and medical care. Morpho will provide broad support, making a wide range of innovations happen with its imaging technology and Deep Learning technology.

Source: Morpho