NVIDIA Grace CPU Conference Sessions

This talk explores how Ansys leveraged NVIDIA Omniverse Blueprint for CAE using CUDA, Modulus, and Omniverse for the creation and deployment of digital twins. We'll dive into developing a high-fidelity CFD solver optimized for NVIDIA hardware (Grace, Hopper, Blackwell), enabling fast and accurate physics simulations. We'll then talk about the Integration of NVIDIA Modulus AI framework to easily build customized AI solutions that can improve designer productivity and quickly identify optimal design configurations. Finally, we'll show how we are leveraging Omniverse for real-time, photo-realistic visualization with examples. This combination of hardware and software solutions empowers engineers to build and deploy highly accurate digital twins with unprecedented speed, dramatically shortening product development cycles.

Important: Near capacity, highly suggest arriving early. Attendees are let in on a first-come, first-served basis.

Engage directly with NVIDIA specialists to discuss the innovative NVIDIA MGX modular reference design. This is a unique opportunity to ask questions, share insights, and gain a deeper understanding of how MGX can revolutionize accelerated computing deployment in data centers. This session is ideal for data center architects, original equipment manufacturers, original design manufacturers, and IT professionals interested in leveraging NVIDIA's cutting-edge technologies to enhance their AI and accelerated computing capabilities. Don't miss this chance to connect with experts and deepen your understanding of NVIDIA's innovative MGX architecture!

Important: Connect with the Experts sessions are interactive sessions that give you a unique opportunity to meet, in either a group or one-on-one setting, with the minds behind NVIDIA’s products and research to get your questions answered. Attendees can meet with experts on a first-come, first-served basis.

Grace, NVIDIA’s first CPU for the data center, packs 72 high-performance Armv9 cores on a single die and integrates LPDDR5 memory for up to 500GB/s of memory bandwidth. Whether paired with NVIDIA’s best-in-class data center GPUs or combined with a second Grace CPU in the Grace Superchip, it delivers exceptional performance. It's designed for HPC and hyperscale data center deployments where absolute performance, energy efficiency, and density matter. In this interactive session, our experts will answer any questions you may have about fully unlocking the potential of the Grace CPU. We welcome anyone interested in evaluating and deploying NVIDIA Grace-based solutions in their data center. You should understand the basics of CPU programming and are encouraged to bring application-specific questions and general questions about the Arm software ecosystem.

Important: Connect with the Experts sessions are interactive sessions that give you a unique opportunity to meet, in either a group or one-on-one setting, with the minds behind NVIDIA’s products and research to get your questions answered. Attendees can meet with experts on a first-come, first-served basis.

Are you looking to accelerate your data analytics workflow on GPUs? Is your workflow bottlenecked by complex and slow-running SQL queries or dataframe operations? Join our session on data analytics workloads on GPUs, where you'll discover how to accelerate your data processing with cutting-edge NVIDIA hardware and software solutions. Learn about libraries like RAPIDS and Spark, powerful tools that leverage GPUs for fast data processing, and explore NVIDIA's Grace Hopper and Grace Blackwell architectures designed to accelerate your data workloads like never before.

Important: Connect with the Experts sessions are interactive sessions that give you a unique opportunity to meet, in either a group or one-on-one setting, with the minds behind NVIDIA’s products and research to get your questions answered. Attendees can meet with experts on a first-come, first-served basis.

RAPIDS open source library suite is the foundation for accelerated AI and data science workloads. It supports dataframes, machine learning, graph analytics, vector databases, and LLM-based workflows. In this session, NVIDIA technical leaders will walk data scientists and engineers through new features, zero code change interfaces, powerful optimizations and give a peek at what’s around the corner.

Important: Near capacity, highly suggest arriving early. Attendees are let in on a first-come, first-served basis.

Talk to experts about programming, code optimization, and topics like device extensions and multi-GPU multi-node support, share your experiences, and get guidance for Math Libraries cuBLAS, cuSOLVER, cuSPARSE, cuSPARSELt, cuDSS, cuTENSOR, cuFFT, cuRAND, the Math API, and NVPL.

Important: Connect with the Experts sessions are interactive sessions that give you a unique opportunity to meet, in either a group or one-on-one setting, with the minds behind NVIDIA’s products and research to get your questions answered. Attendees can meet with experts on a first-come, first-served basis.

Take a deep dive into the latest developments in NVIDIA software for high performance computing applications, including a comprehensive look at what’s new in programming models, compilers, libraries, and tools. We'll cover topics of interest to HPC developers, targeting traditional HPC modeling and simulation, quantum computing, HPC+AI, scientific visualization, and high-performance data analytics.

Important: Near capacity, highly suggest arriving early. Attendees are let in on a first-come, first-served basis.