High Performance Computing

A Supercharged Law

GPU computing is defining a new, supercharged law to replace Moore’s law. It starts with a highly specialized GPU parallel processor and continues through system design, software, algorithms, and optimized applications. Each GPU-accelerated server replaces dozens of commodity CPU servers, delivering a dramatic boost in application throughput and cost savings.

EXPLORE GTC 2019 HPC SESSIONS

LEARN MORE

Accelerating the Rate of Scientific Discovery

The NVIDIA^® CUDA^® programming model is the platform of choice for high-performance application developers, with support for more than 550 GPU-accelerated applications—including the top 15 high performance computing (HPC) applications. From weather prediction and materials science to wind tunnel simulation and genomics, NVIDIA GPU-accelerated computing is at the heart of HPC’s most promising areas of discovery.

Learn more:

> GPU Application Quick Start Guides

> NVIDIA GPU Cloud HPC Application Containers

Powering the World’s Fastest Supercomputers

GPU computing is the most accessible and energy-efficient path forward for HPC and the data center. Today, NVIDIA powers the fastest supercomputers in the U.S. and Europe, as well as some of the most advanced systems under construction.

In the U.S., Oak Ridge National Labs has introduced Summit, the world’s smartest and most powerful supercomputer, with over 200 petaFLOPS for HPC and 3 exaOPS for AI. Summit fuses HPC and AI computing with over 27,000 NVIDIA Volta Tensor Core GPUs to accelerate scientific discovery. And Japan’s AI Bridging Cloud Infrastructure (ABCI) will come online in 2018 as the country’s most powerful supercomputer and a global innovation platform for AI.

Explore the World's Fastest Supercomputer: Summit

Unified Platform for HPC AND AI

The intersection of HPC and AI is extending the reach of science and accelerating the pace of scientific innovation like never before. AI is helping tackle previously unsolvable problems by modeling the world using experimental and simulation data. It’s also helping deliver real-time results with models that used to take days or months to simulate.

Download the InsideBIGDATA Guide to Deep Learning and Artificial Intelligence

Download the Intersection of HPC and Machine Learning Whitepaper

Intersect360: HPC Application Support for GPU Computing

Download Application Workload Technical Overview

HPC AND AI Customer Success Stories

The combination of AI and deep learning algorithms with HPC is expected to profoundly impact every aspect of human life. Here are just a few examples of how that’s already happening:

NCSA Gravity Group

In 2017, the Laser Interferometer Gravitational-Wave Observatory (LIGO) was awarded the Nobel Prize in Physics for detecting gravitational waves millions of light years away in real time.

Seeing Gravity in Real-Time with Deep Learning (PDF 563 KB)

UFL and UNC

The University of Florida (UFL) and the University of North Carolina (UNC) developed the ANAKIN-ME neural network engine to produce computationally fast quantum mechanical simulations with high accuracy at very low cost.

An AL Quantum Breakthrough (PDF 464 KB)

Princeton University: ITER Fusion Energy

Princeton University is leveraging the computational power of GPUs to predict disruptions in a tokamak fusion reactor in ITER, an international experiment seeking to prove the feasibility of fusion as a renewable source of clean energy.

Seeding the Path of Fusion Energy with Deep Learning (PDF 773KB)

Oak Ridge National Laboratory

NVIDIA GPU-powered AI accelerates the mapping and analysis of population distribution around the globe, enabling more efficient planning, delivery of goods and services, and use of scarce resources.

A 21st Century Planning Tool Built on AI (PDF 1.1 MB)

NASA Ames

In order to better keep a finger on the pulse of the Earth’s health, NASA developed DeepSat, a deep learning framework for satellite-image classification and segmentation.

An AL Monitor of Earth's Vitals (PDF 256 KB)

Simplified Programmability

GPUs are at the heart of accelerating HPC, and now, they’re simpler than ever to program with vast libraries, directives-driven OpenACC, and a powerful CUDA programming model.

ACCELERATE THE DATA CENTER.

Learn more about NVIDIA Tesla^® V100, the most advanced data center GPU for AI and HPC.

LEARN MORE