NVIDIA Omniverse can now connect to scientific computing visualisation software and support batch-rendering workloads on systems powered by NVIDIA A100 and H100 Tensor Core GPUs.
With the link, AI and HPC researchers, scientists and engineers can run batch workloads such as videos and images rendering and synthetic 3D data generation.
To foster more seamless, collaborative workflows for the HPC community, NVIDIA has announced connections to popular scientific computing tools such as Kitware’s ParaView, an application for visualisation, NVIDIA IndeX for volumetric rendering, NVIDIA Modulus for developing physics-ML models, and NeuralVDB for large-scale sparse volumetric data representation.
“Today’s scientific computing workflows are extremely complex, involving enormous datasets that are impractical to move and large, global teams that use their own specialised tools. With new support for Omniverse on A100 and H100 systems, HPC customers can finally start to unlock legacy data silos, achieve interoperability in their complex simulation and visualisation pipelines, and generate compelling visuals for their batch-rendering workflows,” said Dion Harris, Lead Product Manager of Accelerated Computing at NVIDIA.
Using Omniverse and hybrid-cloud workloads, scientific computing customers can connect legacy simulation and visualisation pipelines to achieve distributed, fully interactive, true real-time interaction with their models and datasets.
NVIDIA customers such as Argonne National Laboratory, Lockheed Martin and Princeton Plasma Physics Laboratory are already seeing benefits of Omniverse for HPC workloads.
Image courtesy of Prof Tang of Princeton Plasma Physics Laboratory: A digital twin simulation of a fusion device and control system in NVIDIA Omniverse.
Entelechy Asia 