Nvidia introduced its Nvidia Omniverse Blueprint, a know-how that permits trade software program builders to construct digital twins with life like real-time physics.
It allows computer-aided engineering (CAE) clients in aerospace, automotive, manufacturing, power and different industries create digital twins with real-time interactivity. One instance of that’s the digital wind tunnel that automobile designers can use to simulate wind shifting over a simulated automobile.
Nvidia made the bulletins on the SC24 supercomputer occasion.
Software program builders reminiscent of Altair, Ansys, Cadence and Siemens can use the reference workflow, which incorporates Nvidia acceleration libraries, physics-AI frameworks and interactive bodily based mostly rendering, to realize 1,200x quicker simulations and real-time visualization. This helps their clients drive down improvement prices and power utilization whereas attending to market quicker.
“We built Omniverse so that everything can have a digital twin,” stated Jensen Huang, CEO of Nvidia, in a press release. “Omniverse Blueprints are reference pipelines that connect NVIDIA Omniverse with AI technologies, enabling leading CAE software developers to build groundbreaking digital twin workflows that will transform industrial digitalization, from design and manufacturing to operations, for the world’s largest industries.”
One of many first purposes of the blueprint is computational fluid dynamics (CFD) simulations, a crucial step to nearly discover, take a look at and refine the designs of vehicles, airplanes, ships and lots of different merchandise. Conventional engineering workflows — from physics simulation to visualization and design optimization — can take weeks and even months to finish.
In an trade first, Nvidia and Luminary Cloud are demonstrating at SC24 a digital wind tunnel that enables customers to simulate and visualize fluid dynamics at real-time, interactive speeds, even when altering the car mannequin contained in the tunnel.
Unifying three pillars of Nvidia dev tech
Constructing a real-time physics digital twin requires two basic capabilities: actual time physics solver efficiency and real-time visualization of large-scale datasets.
The Omniverse Blueprint achieves these by bringing collectively Nvidia CUDA-X libraries to speed up the solvers, the Nvidia Modulus physics-AI framework to coach and deploy fashions to generate stream fields, and Nvidia Omniverse software programming interfaces for 3D knowledge interoperability and real-time RTX-enabled visualization.
Builders can combine the blueprint as particular person components or in its entirety into their present instruments.
Ansys is the primary buyer to undertake the Omniverse Blueprint, making use of it to Ansys Fluent fluid simulation software program to allow accelerated computational fluid dynamics simulation.
Ansys ran Fluent on the Texas Superior Computing Middle on 320 Nvidia GH200 Grace Hopper superchips.
A 2.5-billion-cell automotive simulation was accomplished in simply over six hours. That may have taken almost a month working on 2,048 x86 CPU cores, considerably enhancing the feasibility of in a single day high-fidelity CFD analyses and establishing a brand new trade benchmark.
“By integrating NVIDIA Omniverse Blueprint with Ansys software, we’re enabling our customers to tackle increasingly complex and detailed simulations more quickly and accurately,” stated Ajei Gopal, CEO of Ansys, in a press release. “Our collaboration is pushing the boundaries of engineering and design across multiple industries.”
Luminary Cloud can be adopting the blueprint. The corporate’s new simulation AI mannequin, constructed on Nvidia Modulus, discovered the relationships between airflow fields and automobile geometry based mostly on coaching knowledge generated from its GPU-accelerated CFD solver.
The mannequin runs simulations orders of magnitude quicker than the solver itself, enabling real-time aerodynamic stream simulation that’s visualized utilizing Omniverse APIs.
Altair, Past Math, Cadence, Hexagon, Neural Idea, Siemens, SimScale and Trane Applied sciences are additionally exploring adoption of the Omniverse Blueprint into their very own purposes.
The Omniverse Blueprint will be run on all main cloud platforms, together with Amazon Net Companies, Google Cloud and Microsoft Azure. It’s also accessible on Nvidia DGX Cloud.
Rescale, a cloud-based platform that helps organizations speed up scientific and engineering breakthroughs, is utilizing the Nvidia Omniverse Blueprint to allow organizations to coach and deploy customized AI fashions in just some clicks.
The Rescale platform automates the complete application-to-hardware stack and will be run throughout any cloud service supplier. Organizations can generate coaching knowledge utilizing any simulation solver; put together, prepare and deploy the AI fashions; run inference predictions; and visualize and optimize fashions.
Corporations thinking about studying extra concerning the Nvidia Omniverse Blueprint for real-time computer-aided engineering digital twins can join early entry.
