The seventh annual GTC kicked off in Silicon Valley today. NVIDIA CEO Jen-Hsun Huang gave the opening keynote for a variety of topics. GPUs have become increasingly significant from video games to Bitcoin mining to supercomputers to VR.
Developers
NVIDIA’s focus for years might have been hardware but now they are emphasizing software to get the most out of their hardware. WITH the CUDA toolkit and other tools for GPU-accelerated computing, the software packages have been bundled into an NVIDIA SDK. This takes advantage of the Pascal architecture and provides the cuDNN 5 library for deep learning, CUDA 8 for accelerated computing, and more elements for self-driving cars, visualization, autonomous machines, gaming, and VR detailed in this post NVIDIA Details Major Software Updates to Sharpen GPU Computing’s Cutting Edge.
Deep Learning
AI was one focus of the GTC 2016 keynote. The NVIDIA DGX-1 is a single integrated system powered by the Tesla P100 chip “with the computing throughput of 250 servers” of 170 teraflops is a deep-learning supercomputer. This system allows training artificial intelligence up to 12 times faster in things such as facial recognition or understanding art.
“Artificial intelligence is the most far-reaching technological advancement in our lifetime,” Jen-Hsun told the audience. “It changes every industry, every company, everything. It will open up markets to benefit everyone. The DGX-1 is easy to deploy and was created for one purpose: to unlock the powers of AI and bring superhuman capabilities to problems that were once unsolvable.”
Massachusetts General Hospital is one of the early customers for the DGX-1. As a clinical data center, the AI will help diagnose disease in the fields of radiology and pathology. The system will learn from the archive of 10 billion medical images held by the nation’s largest research hospital.
VR
VR is a big innovation this year VR viewers and headsets getting a lot of attention. NVIDIA is taking part in this with their Iray technology to create accurate 3D environments that you can explore.
Iray VR capabilities will allow us to create environments that let users strap on a headset and peer around photorealistic virtual environments. With Iray VR, you’ll be able to look around the inside of a virtual car, a modern loft, or the interior of our still unfinished Silicon Valley campus with uncanny accuracy.
As today is the launch day of the HTC Vive, NVIDIA posted about it with their VR-optimized Game Ready drivers and Virtual Reality GeForce Experience Optimal settings on supported games.
Cars
The NVIDIA DRIVE PX 2 can enable autonomous vehicles to have superb perception with a mapping platform provided by NVIDIA Tesla GPUs in the data center. The cars must be able to process inputs from multiple sensors and adapt accordingly. To prove the platform, NVIDIA’s Drive PX 2 AI supercomputer will be powering a car in the ROBORACE Championship, the first global autonomous motor sports competition.
