Cadence enhances Digital Twin Platform with NVIDIA SuperPOD model

Cadence expands its Reality Digital Twin Platform with the latest NVIDIA DGX SuperPOD model, elevating AI factory design and operations.

Cadence has announced an exciting development in its Cadence® Reality™ Digital Twin Platform library - the integration of a digital twin model for the NVIDIA DGX SuperPOD, equipped with DGX GB200 systems. This innovative model is poised to support data centre designers and operators in effortlessly deploying the world's leading AI accelerator in the construction of AI factories.

Cadence's Reality Digital Twin Platform is hailed as a first-of-its-kind solution, enabling users to simulate AI factories and data centres with unparalleled accuracy to meet specific service-level agreements. The Digital Twin allows users to account for key constraints including cost, space, energy, cooling, and environmental factors prior to any physical implementation.

Michael Jackson, Senior Vice President, System Design and Analysis at Cadence, emphasised the importance of such tools in rapidly scaling AI initiatives, stating that meeting design criteria with targeted equipment is crucial for successful outcomes. As such, the introduction of the digital model of NVIDIA’s DGX SuperPOD into the Cadence Reality library marks a significant advancement for mission-critical projects and decision-making.

The platform empowers designers to construct high-fidelity data centre twins by incorporating vendor-provided digital models. These models function akin to their physical versions, allowing engineers to meticulously design data centres and campuses with precise power, space, cooling, and performance parameters. The platform also enables users to explore and react effectively to potential failures and upgrade scenarios.

The partnership between Cadence and NVIDIA is underscored by Tim Costa, NVIDIA's General Manager of Industrial and Computational Engineering, as an essential step towards accelerating AI factory developments. This collaboration is part of a broader strategy to enhance the AI factory build process, with earlier efforts highlighting the platform's integration with NVIDIA's Omniverse blueprint for AI operations.

Maintaining its role as a comprehensive resource, the Reality Digital Twin Platform’s library boasts over 14,000 items from more than 750 vendors, ensuring extensive representation of data centre components. Should a part be unavailable, Cadence commits to adding it upon request as part of its ongoing software support services.

While enriching the library with NVIDIA's SuperPOD model, Cadence remains committed to refining its digital ecosystem to stay aligned with the dynamic requirements of the engineers employing its solutions.

BAC’s immersion cooling solution is now Intel-certified, demonstrating its application in...
Marvell Technology acquires XConn Technologies to strengthen its PCIe and CXL switching portfolio...
Schneider Electric appoints Matthew Baynes as Vice President, driving AI-fueled data centre growth...
Supermicro collaborates with NVIDIA to optimise AI infrastructure delivery with expanded...
NVIDIA unveils AI supercomputers at CES, allowing developers to leverage advanced AI models without...
Accenture is set to acquire DLB Associates to enhance its AI data centre capabilities, aiding...
SUNeVision launches its third Startup Programme, fostering innovation in Hong Kong with robust...
stc group and HUMAIN have formed a joint venture to build AI data centres in Saudi Arabia,...