Presented by Supermicro
Virtual 3D collaboration at scale is here. Learn how NVIDIA’s Omniverse Enterprise platform with the new addition of NVIDIA OVX lets companies across industries build digital twins, and maximize iterations in an easily extensible, open platform, tailored to handle your workload.
Efficient 3D design collaboration between teams is critical across so many industries. But these teams are increasingly dispersed across the globe, working remotely on compute-heavy technologies and incompatible software. That’s where NVIDIA’s Omniverse comes in. It’s a scalable, multi-GPU, real-time reference development platform for 3D simulation and design collaboration through virtual worlds. Teams, assets, and creative applications are united in one interface.
The platform combines NVIDIA’s real-time ray-tracing, physics simulation, and AI technologies with the power of accelerated computing on NVIDIA RTX-enabled workstations and servers to enable remote teams to collaborate live in shared photorealistic 3-D worlds. Omniverse is a scalable platform, not only enabling design collaboration, but also the ability to develop and operate industrial digital twin simulations.
With the recent addition of NVIDIA OVX, which is purpose-built to power large-scale industrial digital twins from the data center, designers, engineers, and planners can build and operate physically accurate digital twins of buildings or create massive, true-to-reality simulated environments with precise time synchronization across physical and virtual worlds.
“NVIDIA is helping users create a digital twin of the world; its partnership with Supermicro means that teams of hundreds of people can collaborate simultaneously to build and manipulate those worlds,” says Erik Grundstrom, Director, FAE & business development director at Supermicro. “The value add of Omniverse Enterprise is not only enabling more efficient collaboration, but time to complete, time to finish your job, time to get things done. Projects today can be accomplished much quicker with Omniverse Enterprise than they could be in the past.”
How Supermicro is super-powering Omniverse use cases
Omniverse Enterprise core collaboration and database engine “Nucleus” enables real-time, instant editing on a shared 3D scene without the need to download and reupload entire scenes and send from designer to designer. Multiple people can then work on these files simultaneously, and make edits in real time. Supermicro ensures that every deployment can handle these real-time edits seamlessly. It designs every customer’s NVIDIA-Certified workstations and servers to match their workload, and not vice versa, and each deployment can be customized to that customer’s particular needs.
“Supermicro has such a vast array of products, so many different SKUs, that we’re able to cater to whatever exactly a customer is looking for, and then accommodate all of the various factors that go into deployment — like GDP, thermals, your power to your rack — and create the best components to deploy this on,” Grundstrom says. “All that you need to run Omniverse Enterprise is an NVIDIA RTX GPU. We can do that in a million different ways, a million different form factors.”
The challenge with around-the-world collaboration is always latency, from the consumer experience back to the virtual desktop infrastructure. But if that problem is solved, then the sky is the limit, Grundstrom says. And with that problem solved, Omniverse Enterprise can accommodate hundreds of users at once. That applies to media and entertainment, digital twin development, architectural design, and so on.
And in media and entertainment, having multiple users with multiple applications working on the same projects simultaneously has huge value. You no longer are waiting for everyone to import and export workflows. Multiple users can also iterate and review on the fly, to improve oversight and guidance, and to speed up approval cycles, making it a huge boost to productivity from a management perspective as well.
The use cases that OVX and digital twins are unlocking
While the primary use cases have primarily been in the industrial manufacturing and warehouse logistics segments, digital twins will enable the next era of industries and AI. A digital twin is a virtual representation of a physical asset, process, or environment that encompasses several autonomous systems all operating at true real time.
In industries where digital simulations have long been hamstrung by cost and computing limitations, digital twin simulation is changing the game with advanced designing, modeling, automation, and testing capabilities. Just as the company has DGX for AI,it now has OVX for Omniverse.
BMW Group is using Omniverse Enterprise capabilities to develop and operate its state-of-the-art factory of the future. BMW engineers can use real-time data and monitoring to push the envelope on smart manufacturing. They are working to optimize the ergonomics for human interactions and increase worker safety and improve product quality while reducing manufacturing costs and unplanned downtime.
Supermicro itself is creating a digital twin of its building in San Jose to emulate 5G RAN for customers. They’re designing and deploying 5G servers for customers to interact with in real time, to monitor sensor data, from temperatures to voltages and more as they run.
But it’s particularly exciting in the science world, Gundstrom says, to be able to log in and collaboratively view, adjust, and optimize data-driven models in real time at the highest possible quality. It takes a lot of compute horsepower to do this, and NVIDIA has solved that problem with Omniverse and OVX. For instance, Omniverse will be able to completely change the way that we visualize things in the medical world since health care delivery and research depend so much on complex, accurate visualization.
“With Omniverse you can really throw as much horsepower at an imaging problem as you want to,” he says. “With multi-GPU render, which is up to 64 GPUs simultaneously, you’re at 100X the performance of the machine in a doctor’s office. To bring that data outside of the doctor’s office and allow a platform like Omniverse to enable multiple-party viewing and interaction, that’s a huge step.”
But the same thing then applies to the sciences unilaterally. Across disciplines, visualization tools are usually quite old, and much of the software they use is not optimized for viewing at higher resolution, or a use case needing a lot of GPU to render something like a 3D model in real time.
“Omniverse changes that for sciences across the board, as well as in manufacturing, media and entertainment, AEC fields, and more,” he says.
Watch on demand to learn:
- How Omniverse Enterprise changes collaboration for creative teams across industries
- The top use cases for 3D design collaboration and industrial digital twins
- How to design and launch custom systems based on your own workloads
- Real-world case studies from NVIDIA’s biggest customers
- Michael Kaplan, Director, Global Business Development, Media & Entertainment at NVIDIA
- Erik Grundstrom, Director, FAE & Business Development, Supermicro
- Charlie Fink, Author and Metaverse Consultant, Moderator