The integration of the NVIDIA Modulus AI framework and the Ansys SeaScape platform enables engineers to easily build customized AI solutions, increasing designer productivity and quickly identifying optimal design configurations. Masu.
/ Main highlights
The NVIDIA Modulus artificial intelligence (AI) framework is integrated into Ansys SeaScape™, a cloud-optimized big data analytics platform for electronic design automation, and has been demonstrated to accelerate thermal simulation by more than 100x.
Modulus’ physics-driven AI technology powers Ansys’ power integrity and reliability signoff platforms, including Ansys RedHawk-SC™, Ansys Totem-SC™, Ansys PathFinder-SC™, and Ansys RedHawk-SC Electrothermal™. A powerful complement to the Ansys multiphysics simulation engine within SeaScape.
This integration improves product outcomes for applications such as graphics processing units (GPUs), high performance computing (HPC) chips, AI chips, smartphone processors, and advanced analog integrated circuits.
pittsburgh, November 19, 2024 /PRNewswire/ — Ansys (NASDAQ: ANSS) today announced the integration of the NVIDIA Modulus AI framework into Ansys semiconductor simulation products, delivering AI capabilities that significantly speed design optimization. This enables engineers to create customized generative AI surrogate models that accelerate design iterations and explore a wider design space. This technology integration will improve outcomes for a wide range of products, including GPUs, HPC chips, AI chips, smartphone processors, and advanced analog integrated circuits.
NVIDIA Modulus is a physical AI framework for training and deploying models that combines physically-based domain knowledge with simulation data, allowing users to create AI engines customized to their needs. As AI is integrated into computer-aided engineering workflows, it is important for users to have a seamless, integrated pipeline that allows data generated by solvers to flow into the AI framework used to train models. That becomes important. By integrating the NVIDIA Modulus framework into the Ansys SeaScape platform, customers can train their AI engines using high-fidelity data generated by Ansys tools and use the newly created engine to create more robust designs. You will be able to consider.
For example, designers can use Ansys RedHawk-SC’s library of completed designs to train AI models with the integrated Modulus framework. Once the AI is trained, it can be used to quickly identify the best design based on desired specifications such as size, power, and performance. Ansys plans to add AI accelerators created by Modulus to semiconductor solutions such as RedHawk-SC, Totem-SC, PathFinder-SC, and RedHawk-SC Electrothermal for faster thermal simulations and easier power calculations is. With this AI-enhanced process, Ansys and NVIDIA have demonstrated more than 100x speedup for thermal simulations.
story continues
