Boosting Engineering Design Productivity with NVIDIA Grace Blackwell and AI

Scientific discovery is at the heart of human progress, unveiling new possibilities and solving the world's most complex challenges. Advanced electronics play a pivotal role in this process, serving as the foundation for groundbreaking technologies across healthcare, energy, and communications. Yet, as systems become more sophisticated, the design process grows increasingly intricate, demanding unprecedented precision and quality. Meanwhile, the pursuit of innovation opens endless opportunities, but it also imposes tighter project timelines and greater pressure to deliver rapid results without compromising excellence.

To tackle these challenges, Cadence and NVIDIA have teamed up to lead this transformation, combining NVIDIA's powerful accelerated computing platforms with Cadence's expertise in engineering design. Together, we're reshaping workflows, enhancing productivity, and enabling industries to push the boundaries of what's possible. Our collaboration supports rapid innovation, stronger sustainability, and game-changing advances across design and scientific research.

Accelerating Scientific Computing with NVIDIA Grace Blackwell

Figure 1: Accelerating Science and Engineering on NVIDIA Blackwell Platform

Cadence is accelerating scientific innovation by leveraging the NVIDIA Blackwell platform. Cadence solutions span a broad spectrum of design aspects, from core EDA tools to system design and multiphysics analysis engines and digital biology. Cadence EDA and hardware verification tools played a key role in the creation of the Blackwell platform. Together, they can deliver significant computational gains in throughput, increased cost efficiency, and improved sustainability.

Recently, we co-optimized the Cadence solver engines with the NVIDIA Blackwell accelerated computing platform, resulting in up to 80X reduced computational fluid dynamics (CFD) simulation time, up to 10X accelerated SPICE simulation time, and 3D-IC design and analysis for thermal, stress, and warpage accelerated by up to 7X when compared to X86 platforms.

These results help solve one of the grand challenges of CFD—a simulation of the complete aircraft within the most challenging part of the flight envelope—takeoff and landing. Using the Cadence Fidelity CFD Platform, we successfully ran multi-billion cell simulations on NVIDIA Grace Blackwell GB200 GPUs in under 24 hours that would previously have required an entire top 500 CPU cluster with 100,000s of cores and lasting several days to complete. Cadence will continue to leverage Blackwell to test the limits of the simulation and, as a result, help the aerospace industry reduce the amount of wind tunnel tests while reducing cost and expediting time, thereby saving cost and speeding time to market.

Figure 2: CAE Acceleration with NVIDIA Grace Blackwell

Advances in AI Factory Design

In collaboration with NVIDIA, Cadence is transforming AI factory design by enabling faster creation of high-quality, real-time digital twins—virtual representations crucial for modeling, analyzing, and optimizing AI factories with sustainability and efficiency in mind. At the core of this initiative is the integration of the Cadence Reality Digital Twin Platform application with the NVIDIA Omniverse Blueprint infrastructure for real-time computer-aided engineering. Cadence is one of the first adopters of the blueprint, which defines best practices and provides reference examples for data center engineers to design, simulate, and accelerate the development of new AI factories.

A key advancement is the definition of a standard of digital twin component modeling for AI data center design. By adopting Universal Scene Description (OpenUSD), the industry can deliver interoperable models, simplifying the rapid assembly of high-quality AI factory digital twins with components provided by participating electronics and equipment suppliers. Cadence has joined the Alliance for Open Universal Scene Description (AOUSD) to contribute to the definition of this standard and is chairing an interest group focused on defining digital twins.

Figure 3: Alliance for Open USD for Modeling

The resulting digital twins enable engineering teams to design and visualize AI factories with unprecedented accuracy and detail. The Cadence Reality Digital Twin Platform enhances these capabilities with next-generation, physically based simulations that allow immersive and dynamic modeling of data center components.

Figure 4: Omniverse Blueprint for Real-Time CAE Digital Twins

Another pivotal development is the seamless integration of NVIDIA Omniverse Viewport, Cadence Allegro X Design Platform, and the Cadence Reality Digital Twin Platform. These Viewports provide designers with more accurate visualization for the entire electronic system design process, enabling intuitive creation, integration, and optimization of the systems.

Figure 5: Digital Twins for Data Centers

Full-Stack Agentic AI Engineering Solution

Our next phase of innovation focuses on a full-stack agentic AI reasoning solution for engineering design and science discovery, providing significant designer productivity with AI and human-guided design refinement. Cadence is collaborating with NVIDIA to bring agentic AI reasoning to the entire SoC design process (IP creation and SoC integration), integrating the Cadence Joint Enterprise Data and AI (JedAI) Platform with NVIDIA NeMo, part of the NVIDIA AI Enterprise software platform, to create the most advanced agentic AI infrastructure. In addition, Cadence is working with NVIDIA to integrate their new open Llama Nemotron AI models with advanced reasoning capabilities to develop "silicon agents" to accelerate designer productivity for tasks such as:

• Intelligent conversational AI assistants for boosting user productivity and innovation
• Deep reasoning for verification based on the underlying design collateral and verification agents
• Design generation and optimization with design agents for digital and custom circuits

Figure 6: Cadence Agentic AI Stack

Additionally, we are extending the AI paradigm to the realm of life sciences by integrating our cloud-native Orion Molecular Design Platform with NVIDIA BioNeMo NIM microservices. This scientific software and hardware development collaboration will produce transformative tools for drug discovery. NIM microservices are supported with unprecedented on-demand and reserved GPU access through the Orion platform, allowing these calculations to be performed at an enormous scale and rapidly providing new hypotheses for downstream tools. This combination expands capabilities in AI models for 3D de novo protein structure prediction, small molecule generative AI, and customized AI models for molecular property prediction. Candidate molecules generated with NIM microservices can be iteratively designed against and improved upon with Orion tools, including physically rigorous molecular dynamics simulations.

Figure 7: Cadence Orion Platform with NVIDIA BioNeMo NIM Microservices

MediaTek 30% Productivity Gain for 2nm Development

An example of how modern accelerated compute and AI can help joint customers is seen with MediaTek. MediaTek has adopted the AI-driven Cadence Virtuoso Studio and Spectre X Simulator on the NVIDIA accelerated computing platform for its 2nm development. As design size and complexity continue to escalate, advanced-node technology development has become increasingly challenging for SoC providers. To meet the aggressive performance and turnaround time (TAT) requirements for its 2nm high-speed analog IP, MediaTek is leveraging Cadence's proven custom/analog design solutions, enhanced by AI, to achieve a 30% productivity gain.

Figure 8: MediaTek Adopts AI-Driven Design Solutions on NVIDIA Platform

Read how MediaTek Adopts AI-Driven Cadence Virtuoso Studio and Spectre Simulation on NVIDIA Accelerated Computing Platform for 2nm Designs:

"As MediaTek continues to push technology boundaries for 2nm development, we need a trusted design solution with strong AI-powered tools to achieve our goals," said Ching San Wu, corporate vice president at MediaTek. "Closely collaborating with Cadence, we have adopted the Cadence Virtuoso Studio and Spectre X Simulator, which deliver the performance and accuracy necessary to achieve our tight design turnaround time requirements. Cadence's comprehensive automation features enhance our throughput and efficiency, enabling our designers to be 30% more productive."

Conclusion

Figure 9: Cadence and NVIDIA collaboration

Our collaboration with NVIDIA is the driving force in the advancement of Intelligent System Design. By leveraging AI combined with accelerated compute, we are enabling significant productivity gains across semiconductor design, AI factory development, and life sciences. We are not only solving today's complex challenges but also setting the stage for next-generation solutions that prioritize sustainability, precision, and scalability.

Read more about Cadence's collaboration with NVIDIA.

See us at GTC!