NVIDIA DGX SATURNV Ranked World’s Most Efficient Supercomputer by Wide Margin

Already speeding our efforts to build smarter cars and more powerful GPUs, NVIDIA’s new DGX SATURNV supercomputer is ranked the world’s most efficient — and 28th fastest overall — on the Top500 list of supercomputers released Monday.

Our SATURNV supercomputer, powered by new Tesla P100 GPUs, delivers 9.46 gigaflops/watt — a 42 percent improvement from the 6.67 gigaflops/watt delivered by the most efficient machine on the Top500 list released just last June. Compared with a supercomputer of similar performance, the Camphore 2 system, which is powered by Xeon Phi Knights Landing, SATURNV is 2.3x more energy efficient.

That efficiency is key to building machines capable of reaching exascale speeds — that’s 1 quintillion, or 1 billion billion, floating-point operations per second. Such a machine could help design efficient new combustion engines, model clean-burning fusion reactors and achieve new breakthroughs in medical research.

GPUs — with their massively parallel architecture — have long powered some of the world’s fastest supercomputers. More recently, they’ve been key to an AI boom that’s given us machines that perceive the world as we do, understand our language and learn from examples in ways that exceed our own (see “Accelerating AI with GPUs: A New Computing Model”).

Using GPUs to Design GPUs

We’re convinced AI can give every company a competitive advantage. That’s why we’ve assembled the world’s most efficient — and one of the most powerful — supercomputers to aid us in our own work.

Assembled by a team of a dozen engineers using 125 DGX-1s — the AI supercomputer in a box we unveiled in April — SATURNV helps us build the autonomous driving software that’s a key part of our NVIDIA DRIVE PX 2 self-driving vehicle platform.

We’re also training neural networks to understand chipset design and very-large-scale-integration, so our engineers can work more quickly and efficiently. Yes, we’re using GPUs to help us design GPUs.

Most importantly, SATURNV’s power will give us the ability to train — and design — new deep learning networks quickly.

DGX-1: Where AI and Supercomputing Intersect

We think such systems can unlock the power of AI for enterprises, research groups and academia.

DGX-1 is an appliance that integrates deep learning software, development tools and eight of our Tesla P100 GPUs — based on our new Pascal architecture — to pack computing power equal to 250 x86 servers into a device about the size of a stove top.

Since then, DGX-1 has been adopted by teams looking to harness AI in a wide variety of settings.

• Enterprise software giant SAP is using DGX-1 AI supercomputers to build machine learning solutions for its 320,000 customers.

• Researchers at groups such as OpenAI, Stanford and New York University are using DGX-1 for their cutting-edge work.

• Startup BenevolentAI uses DGX-1 as part of its effort to accelerate drug discovery by using deep natural language processing, machine learning and AI to formulate new, usable knowledge from complex scientific information.

We’re confident AI — and DGX-1  — will play a key role in even more breakthroughs to come.

The post NVIDIA DGX SATURNV Ranked World’s Most Efficient Supercomputer by Wide Margin appeared first on The Official NVIDIA Blog.

by Roy Kim via The Official NVIDIA BlogThe Official NVIDIA Blog

Get Smart: Join the 7,000+ People Trained on Deep Learning This Year

When the Deep Learning Institute was conceived, we had a hunch that the growing interest in the topic would result in big demand for training in this powerful new technology.

Turns out, it’s even bigger than we expected.

In just nine months, the institute has trained over 7,000 people worldwide in how to apply deep learning to solve complex problems in such areas as image recognition, object detection and natural language processing.

Next week at SC16, North America’s biggest supercomputing tradeshow, the institute will offer a workshop with two hands-on labs:

• Getting Started with Deep Learning will introduce machine learning workflows and provide hands-on experience using deep neural networks to solve a real-world image classification problem.
• Deep Learning for Image Segmentation will cover training and evaluating image segmentation networks, using a medical imaging dataset as an example.

If you’ll be in Salt Lake City for the conference, register now for the Deep Learning Workshops.

Hands-On Instruction

More than 18,000 hours of hands-on instruction have been delivered to data scientists and developers during our global series of GPU Technology Conferences (GTCs) in seven cities worldwide, stretching from Amsterdam to Beijing.

These numbers indicate the growing interest in deep learning as it transforms industries such as automotive, retail and medicine. Deep learning is at work in facial and speech recognition and even on popular websites such as Pinterest and Etsy, where deep learning is used for recommendation engines. Deep learning can even help you lose weight. Breakthroughs in this burgeoning field are driving the expansion of modern AI.

The mission of the Deep Learning Institute is to deliver hands-on training for developers and data scientists that helps them apply deep learning to solve challenging problems on NVIDIA’s AI computing platform.

The institute works with Udacity, Microsoft, Coursera and other educational partners. Classes are available online and as onsite workshops and are suitable for all skill levels. They range from the basics of deep learning to how to build a self-driving car.

In addition to SC16, upcoming Deep Learning Institute onsite classes include GTCx India in Mumbai on Dec. 6. Check out the full schedule, and learn more.

The post Get Smart: Join the 7,000+ People Trained on Deep Learning This Year appeared first on The Official NVIDIA Blog.

by Will Ramey via The Official NVIDIA BlogThe Official NVIDIA Blog

We’re Bringing Our Startup Event to the Startup Nation

Startups need to move fast. GPUs help them move fast. Israel is full of startups.

Makes sense, then, that we’re taking our Emerging Companies Summit to Israel next week, where we’ll bring together more than 85 experts in AI.

Execs from about three dozen young, local companies involved in visual computing, high performance computing and computer vision will be at the heart of the get-together Wednesday, Nov. 16 – our fourth annual edition of the event.

It’s part of our global effort to help nurture startups that are using GPUs in innovative ways.

Israel, of course, is a can’t-miss destination on any tour of the world’s startups. There are 5,000 startups in this country of 8 million people, and Israel has attracted twice as much venture capital investment per capita as the U.S. and more than 30 times that of Europe.

These startups — along with the region’s top-flight universities — are a key reason we opened our first office in Israel last month, staffed with a dozen engineers. Think of it as a startup of our own.

Get more details on Emerging Companies Summit Israel.

The post We’re Bringing Our Startup Event to the Startup Nation appeared first on The Official NVIDIA Blog.

by Alain Tiquet via The Official NVIDIA BlogThe Official NVIDIA Blog