The CUDA platform is the foundation of the GPU computing ecosystem. Every application and framework that uses the GPU does so through CUDA's libraries, compilers, runtimes and language — which means CUDA is growing as fast as its ecosystem is evolving. At this engineering-focused talk, you'll learn from…

This talk is the first part in a series of Core Performance optimization techniques. It is intended for developers learning CUDA, and will teach all the basics that every CUDA developer should know to achieve good performance when writing CUDA kernels for NVIDIA GPUs. The topics covered will include …

Join one of CUDA's architects in a deep dive into how to map an application onto a massively parallel machine, covering a range of different techniques aimed at getting the most out of the GPU. We'll cover principles of parallel program design and connect them to details of GPU programming to see how …

As the raw compute FLOPS become faster and memory bandwidth becomes higher for the latest GPUs, it becomes challenging for applications that launch large numbers of lightweight kernels to saturate GPU compute resources. We'll present the challenges we faced when adapting Desmond, the state-of-art …

NVIDIA’s GPU-accelerated Math Libraries, which are part of the CUDA Toolkit and the HPC SDK, are constantly expanding, providing industry-leading performance and coverage of common compute workflows across AI, ML, and HPC. We'll do a deep dive into some of the latest advancements in the …

Learn how to efficiently manage heterogeneous GPU memory in complex CUDA Python and C++ workflows. Modern AI and analytics use a variety of libraries to process massive data. Careful memory management throughout the workflow is essential to prevent libraries from competing for scarce …

NVIDIA’s H100 introduced fourth-generation Tensor Cores to GPU computing, with over twice the peak performance of the previous generation. This session will build on our GTC’23 session. We'll describe how the latest version of CUTLASS leverages Hopper features for peak performance, covering major new …

ISO-standard languages are the gold standard for productive, portable code, whether programming for a CPU or GPU. See the state of the art in parallel programming with standard language features like the C++ parallel algorithms library and Fortran Do Concurrent. Learn how these languages enable …

This talk is the second part in a series of Core Performance optimization techniques. It is intended for developers who are already familiar with the basics covered in the first part. We'll teach advanced techniques, and how to use some of the new features introduced in Hoppper. The topics covered will …

Do you need to compute larger or faster than a single GPU allows? Learn how to scale your application to multiple GPUs and multiple nodes. We'll explain how to use the different available multi-GPU programming models and describe their individual advantages. All programming models, including …

A paradigm shift from software-defined vehicles to AI-defined vehicles is underway. By using less code, larger models, more compute, and bigger data, NVIDIA is leading the AI-centric approach to AV 2.0. We'll showcase how the end-to-end NVIDIA DRIVE platform—with co-developed hardware, software, and …

Explore the critical role of high-performance, energy-efficient AI computing in developing autonomous vehicles. Learn how NVIDIA DGX Cloud enables scalable computing that facilitates efficient AI model training and supports real-time decision making. NVIDIA AI Enterprise provides developers with …