In 2006, NVIDIA introduced CUDA, a parallel computing platform and programming model, marking the beginning of a new era in computing. Since then, NVIDIA has been at the forefront of advancements in artificial intelligence (AI) and accelerated computing. The latest TOP500 list, which ranks the world’s most powerful supercomputers, underscores the significant achievements NVIDIA has made in these fields.
This year, 384 of the systems on the TOP500 list are powered by NVIDIA technologies. Among the 53 new entries on the list, 87% are accelerated systems. Of these accelerated systems, 85% utilize NVIDIA’s Hopper GPUs, which are instrumental in driving advancements in numerous fields such as climate forecasting, drug discovery, and quantum simulation.
Supercomputing today goes beyond just measuring floating point operations per second (FLOPS). It demands application-specific optimization across the entire computing stack. At the SC24 conference, NVIDIA introduced cuPyNumeric, an NVIDIA CUDA-X library. This library allows over 5 million developers to scale their Python code effortlessly to powerful computing clusters without requiring modifications.
Moreover, NVIDIA announced major updates to the NVIDIA CUDA-Q development platform. This platform empowers quantum researchers by enabling them to simulate quantum devices at unprecedented scales, which was previously deemed computationally impossible.
NVIDIA’s commitment to innovation and excellence has not gone unnoticed. The company has been honored with nearly a dozen HPCwire Readers’ and Editors’ Choice awards in various categories, marking its 20th consecutive year of recognition.
A New Era of Scientific Discovery with Mixed Precision and AI
Mixed-precision floating-point operations and AI are becoming essential tools for researchers tackling the complexities of modern science. These methods provide greater speed, efficiency, and adaptability compared to traditional approaches, without sacrificing accuracy.
This transformation is already underway. During SC24, two projects that were finalists for the Gordon Bell Prize demonstrated how AI and mixed precision are being used to advance genomics and protein design. David Keyes, a professor at King Abdullah University of Science and Technology, employed 0.8 exaflops of mixed precision to explore the connections between genomes, their genotypes, and the prevalence of diseases they are susceptible to.
Similarly, Arvind Ramanathan, a computational biologist at Argonne National Laboratory, utilized 3 exaflops of AI performance on the NVIDIA Grace Hopper-powered Alps system to accelerate protein design. To further enhance AI-driven drug discovery and the development of lifesaving therapies, researchers can use NVIDIA BioNeMo, a suite of tools specifically designed for pharmaceutical applications. The BioNeMo Framework, now open source, facilitates the creation, customization, and deployment of AI models for drug discovery and molecular design.
Across the TOP500 list, the widespread adoption of AI and mixed-precision floating-point operations reflects a global shift in computing priorities. With a total of 249 exaflops of AI performance available, TOP500 systems are propelling innovations and discoveries across various industries.
NVIDIA-accelerated systems on the TOP500 list excel in key metrics such as AI and mixed-precision system performance. With over 190 exaflops of AI performance and 17 exaflops of single-precision (FP32) performance, NVIDIA’s accelerated computing platform has become the engine driving scientific computing. Additionally, the company provides 4 exaflops of double-precision (FP64) performance for specific scientific calculations that require it.
Accelerated Computing is Sustainable Computing
As the demand for computational power increases, so does the need for sustainable solutions. In the Green500 list, which ranks the world’s most energy-efficient supercomputers, systems equipped with NVIDIA’s accelerated computing technologies rank among the top performers, with eight of the top 10 positions. For instance, the JEDI system at EuroHPC/FZJ achieves an impressive 72.7 gigaflops per watt, setting a new benchmark for performance aligned with sustainability.
In the realm of climate forecasting, NVIDIA introduced two new NIM (NVIDIA Intelligent Microservices) microservices for NVIDIA Earth-2, a digital twin platform for simulating and visualizing weather and climate conditions. The CorrDiff NIM and FourCastNet NIM microservices can accelerate climate change modeling and simulation results by up to 500 times.
In a world increasingly aware of its environmental footprint, NVIDIA’s innovations in accelerated computing strike a balance between high performance and energy efficiency, paving the way for a more sustainable future.
Supercomputing Community Embraces NVIDIA
The 11 HPCwire Readers’ Choice and Editors’ Choice awards that NVIDIA received are a testament to the collaborative efforts of the entire scientific community, including engineers, developers, researchers, partners, and customers. The awards include:
- Readers’ Choice: Best AI Product or Technology – NVIDIA GH200 Grace Hopper Superchip
- Readers’ Choice: Best HPC Interconnect Product or Technology – NVIDIA Quantum-X800
- Readers’ Choice: Best HPC Server Product or Technology – NVIDIA Grace CPU Superchip
- Readers’ Choice: Top 5 New Products or Technologies to Watch – NVIDIA Quantum-X800
- Readers’ Choice: Top 5 New Products or Technologies to Watch – NVIDIA Spectrum-X
- Readers’ and Editors’ Choice: Top 5 New Products or Technologies to Watch – NVIDIA Blackwell GPU
- Editors’ Choice: Top 5 New Products or Technologies to Watch – NVIDIA CUDA-Q
- Readers’ Choice: Top 5 Vendors to Watch – NVIDIA
- Readers’ Choice: Best HPC Response to Societal Plight – NVIDIA Earth-2
- Editors’ Choice: Best Use of HPC in Energy – Real-time simulation of CO2 plume migration in carbon capture and storage
- Readers’ Choice Award: Best HPC Collaboration – National Artificial Intelligence Research Resource Pilot
For those interested in learning more about NVIDIA’s latest developments in supercomputing, the company’s special address at SC24 can be watched online. Additionally, more details can be found in the SC24 online press kit.
For further information on NVIDIA’s software products, please refer to their legal information page.
For more details on this topic, you can visit the original NVIDIA blog post.
For more Information, Refer to this article.
