Nvidia announced that Nvidia Blackwell {hardware} will speed up by 50 instances the large computer-aided software program engineering corporations’ software program for digital twins.
The distributors embody Ansys, Altair, Cadence, Siemens and Synopsys. With such accelerated software program, together with Nvidia CUDA-X libraries and blueprints to optimize efficiency reminiscent of automotive, aerospace, power, manufacturing and life sciences can considerably cut back product improvement time, minimize prices and improve design accuracy whereas sustaining power effectivity.
“CUDA-accelerated bodily simulation on Nvidia Blackwell has enhanced real-time digital twins and is reimagining your entire engineering course of,” mentioned Jensen Huang, founder and CEO of Nvidia, in a press release. “The day is coming when just about all merchandise will probably be created and delivered to life as a digital twin lengthy earlier than it’s realized bodily.”
The corporate unveiled the information throughout Huang’s keynote on the GTC 2025 occasion. As famous in my current Q&A with Ansys CTO Prith Banerjee, the hole between simulation and actuality is closing, not simply in issues like video video games but in addition with the same know-how used for engineering simulations.
Ecosystem help for Nvidia Blackwell
Software program suppliers may also help their prospects develop digital twins with real-time interactivity and now speed up them with NVIDIA Blackwell applied sciences.
The rising ecosystem integrating Blackwell into its software program consists of Altair, Ansys, BeyondMath, Cadence, COMSOL, ENGYS, Flexcompute, Hexagon, Luminary Cloud, M-Star, NAVASTO, an Autodesk firm, Neural Idea, nTop, Rescale, Siemens, Simscale, Synopsys and Volcano Platforms.
Cadence is utilizing Nvidia Grace Blackwell-accelerated methods to assist resolve one in every of computational fluid dynamics’ largest challenges — the simulation of a whole plane throughout takeoff and touchdown. Utilizing the Cadence Constancy CFD solver, Cadence efficiently ran multibillion cell simulations on a single NVIDIA GB200 NVL72 server in underneath 24 hours, which might have beforehand required a CPU cluster with a whole lot of 1000’s of cores and a number of other days to finish.
This breakthrough will assist the aerospace trade transfer towards designing safer, extra environment friendly aircrafts whereas decreasing the quantity of pricey wind-tunnel testing required, rushing time to market.
Anirudh Devgan, president and CEO of Cadence, mentioned in a press release, “Nvidia Blackwell’s acceleration of
the Cadence.AI portfolio delivers elevated productiveness and high quality of outcomes for clever system design — decreasing engineering duties that took hours to minutes and unlocking simulations not attainable earlier than. Our collaboration with Nvidia drives innovation throughout semiconductors, knowledge facilities, bodily AI and sciences.”
Sassine Ghazi, president and CEO of Synopsys, mentioned, “At GTC, we’re unveiling the most recent efficiency outcomes noticed throughout our main portfolio when optimizing Synopsys options for NVIDIA Blackwell to speed up computationally intensive chip design workflows. Synopsys know-how is mission-critical to the productiveness and capabilities of engineering groups, from silicon to methods. By harnessing the facility of Nvidia accelerated computing, we may also help prospects unlock new ranges of efficiency and ship their improvements even sooner.”
Ajei Gopal, president and CEO of Ansys, mentioned, in a press release: “The shut collaboration between Ansys and NVIDIA is accelerating innovation at an unprecedented tempo. By harnessing the computational efficiency of NVIDIA Blackwell GPUs, we at Ansys are empowering engineers at Volvo Vehicles to sort out essentially the most advanced computational fluid dynamics challenges with distinctive velocity and accuracy — enabling extra optimization research and delivering extra performant autos.”
James Scapa, founder and CEO of Altair, mentioned in a press release, “The Nvidia Blackwell platform’s computing energy, mixed with Altair’s cutting-edge simulation instruments, provides customers transformative capabilities. This mixture makes GPU-based simulations as much as 1.6 instances sooner in contrast with the earlier era, serving to engineers quickly resolve design challenges and giving industries the facility to create safer, extra sustainable merchandise by real-time digital twins and physics-informed AI.”
Roland Busch, President and CEO of Siemens mentioned: “The mix of Nvidia’s groundbreaking Blackwell structure with Siemens’ physics-based digital twins will allow engineers to drastically cut back improvement instances and prices by utilizing photo-realistic, interactive digital twins. This collaboration will enable us to assist prospects like BMW innovate sooner, optimize processes, and obtain outstanding ranges of effectivity in design and manufacturing.”
Rescale CAE Hub With Nvidia Blackwell
Rescale’s newly launched CAE Hub allows prospects to streamline their entry to Nvidia applied sciences and CUDA®-accelerated software program developed by main unbiased software program distributors. Rescale CAE Hub gives versatile, high-performance computing and AI applied sciences within the cloud powered by Nvidia GPUs and Nvidia DGX Cloud.
Growth Supersonic, the corporate constructing the world’s quickest airliner, will use the Nvidia Omniverse Blueprint for real-time digital twins and Blackwell-accelerated CFD solvers on Rescale CAE Hub to design and optimize its new supersonic passenger jet.
The corporate’s product improvement cycle, which is nearly totally simulation-driven, will use the Rescale platform accelerated by Blackwell GPUs to check totally different flight situations and refine necessities in a steady loop with simulation.
The adoption of the Rescale CAE Hub powered by Blackwell GPUs expands Growth Supersonic’s collaboration with Nvidia. By the Nvidia PhysicsNeMo framework and the Rescale AI Physics platform, Growth Supersonic can unlock 4x extra design explorations for its supersonic airliner, rushing iteration to enhance efficiency and time to market.
The Nvidia Omniverse Blueprint for real-time digital twins, now typically out there, can be a part of the Rescale CAE Hub. The blueprint brings collectively Nvidia CUDA-X libraries, Nvidia PhysicsNeMo AI and the Nvidia Omniverse platform — and can be including the primary Nvidia NIM microservice for exterior aerodynamics, the research of how air strikes round objects.
