Digitalization that mixes AI and simulation is redefining how industrial merchandise are created and remodeling how folks work together with the digital world.
To assist enterprises deal with advanced new workloads, NVIDIA has unveiled the third era of its NVIDIA OVX computing system.
OVX is designed to energy large-scale digital twins constructed on NVIDIA Omniverse Enterprise, a platform for creating and working metaverse purposes. The newest OVX system gives the breakthrough graphics and AI required to speed up large digital twin simulations and different demanding purposes by combining NVIDIA BlueField-3 DPUs with NVIDIA L40 GPUs, ConnectX-7 SmartNICs and the NVIDIA Spectrum Ethernet platform.
A few of the world’s largest methods makers will likely be bringing the most recent OVX methods to prospects worldwide later this yr, offering enterprises with the know-how to deal with advanced manufacturing, design and Omniverse-based workloads. Companies can benefit from the real-time, true-to-reality capabilities of OVX to collaborate on essentially the most difficult visualization, digital workstation and knowledge heart processing workflows.
Reimagining Digital Twin Simulation
Prospects utilizing third-generation OVX methods can velocity their workflows and optimize simulations by immersive digital twins used to mannequin factories, cities, autonomous automobiles and extra earlier than deployment in the actual world. This helps maximize operational effectivity and predictive planning capabilities.
For instance, DB Netze’s Digitale Schiene Deutschland is leveraging the capabilities of OVX to energy large-scale digital twins of dynamic bodily methods, together with rail networks. Others, like Jaguar Land Rover, are leveraging the graphics and simulation capabilities of OVX methods at the side of the NVIDIA DRIVE Sim platform to speed up the testing and growth of next-generation autonomous automobiles.
Subsequent-Era Platform Options
The third era of OVX contains a new structure, with a server design primarily based on a dual-CPU platform with 4 NVIDIA L40 GPUs. Based mostly on the Ada Lovelace structure, the L40 GPU delivers revolutionary neural graphics, AI compute and the efficiency wanted for essentially the most demanding Omniverse workloads.
Every OVX server additionally consists of two high-performance ConnectX-7 SmartNICs to allow multi-node scalability and exact time synchronization. The Ethernet adapters allow the multi-node scalability of OVX methods and supply networking capabilities for the low-latency, high-bandwidth communication that globally dispersed groups want.
New with this era, the BlueField-3 knowledge processing unit offloads, accelerates and isolates CPU-intensive infrastructure duties. For deploying Omniverse at knowledge heart scale, BlueField-3 DPUs present a safe basis for operating the information heart control-plane, enabling greater efficiency, limitless scaling, zero-trust safety and higher economics.
Serving to customers sustain with networking efficiency, the accelerated NVIDIA Spectrum Ethernet platform gives excessive bandwidth and community synchronization to boost real-time simulation capabilities.
Availability
Along with unique NVIDIA OVX companions Lenovo and Supermicro, third-generation OVX methods will likely be accessible later this yr by Dell Applied sciences, GIGABYTE and QCT. NVIDIA can also be engaged on Digital Twin as a Service choices primarily based on OVX with HPE Greenlake.
To be taught extra about OVX, watch NVIDIA founder and CEO Jensen Huang’s GTC keynote.
Register free for NVIDIA GTC, a world AI convention, to attend periods with NVIDIA and business leaders:
