Deep Learning and Artificial Intelligence was NVIDIA’s core focus at GTC 2016. NVIDIA® launched the NVIDIA® DGX-1, the World’s First Deep Learning Supercomputer (available exclusively through XENON) which is a fully integrated hardware and software solution that can be deployed quickly and easily.
The NVIDIA® DGX-1 is built with ground-breaking new NVIDIA® Pascal-powered Tesla P100 accelerators, interconnected with NVIDIA® NVLink™. Its revolutionary performance significantly accelerates training time and most major frameworks making the NVIDIA® DGX-1 the world’s first deep learning supercomputer in a box.
The NVIDIA® DGX-1 system includes a complete suite of optimised deep learning software that allows researchers and data scientists to quickly and easily train deep neural networks. The DGX-1 software includes the NVIDIA® Deep Learning GPU Training System (DIGITS™), a complete, interactive system for designing deep neural networks (DNNs). It also includes the newly released NVIDIA® CUDA® Deep Neural Network library (cuDNN) version 5, a GPU-accelerated library of primitives for designing DNNs.
It also includes optimized versions of several widely used deep learning frameworks — Caffe, Theano and Torch. The DGX-1 additionally provides access to cloud management tools, software updates and a repository for containerized applications.
The NVIDIA® DGX-1 system specifications include:
- Up to 170 teraflops of half-precision (FP16) peak performance
- Eight Tesla P100 GPU accelerators, 16GB memory per GPU
- NVLink Hybrid Cube Mesh
- 7TB SSD DL Cache
- Dual 10GbE, Quad InfiniBand 100Gb networking
- 3U – 3200W
XENON® with NVIDIA® have also developed their own Deep Learning solution the DEVCUBE which contains four TITAN X GPUs with 7 TFlops of single precision, 336.5GB/s of memory bandwidth with 12 GB of memory per board.
XENON’s DEVCUBE also features:
- NVIDIA® DIGITS software providing powerful design, training, and visualisation
- Pre-installed standard Ubuntu 14.04 w/ Caffe, Torch, Theano, BIDMach, cuDNN v2, and CUDA 7.0
- A single deskside machine that plugs into standard wall plug socket
All this results into:
- faster turnaround times
- the freedom to explore multiple network architectures
- accelerated dataset manipulation
- an all in one powerful, energy-efficient, cool, and quiet solution
XENON® DEVCUBE specifications:
- Intel® X99 Platform
- Intel® i7-5960X/3.0GHz/8Cores/20MB/LGA2011
- All in One Liquid Cooling
- 64GB DDR4-2400 Memory
- Up to 4x GTX 980/Titan X GPU’s
- 1x 240GB SSD
- 1x 480GB SSD
- 3x 3TB 3.5” SATA 7200RPM 64MB Enterprise Drives
- Core X9 CUBE Chassis with 1500W PS
XENON’s DEVCUBE is available now.
DIGITS software package is now available in version 3
NVIDIA® Jetson TX1 Developer Kit – Now Available
NVIDIA® Jetson is the world’s leading visual computing platform for GPU-accelerated parallel processing in the mobile embedded systems market. Its high-performance, low-energy computing for deep learning and computer vision makes Jetson the ideal solution for compute-intensive embedded projects.
Other Major Announcements
- NVIDIA® NVIDIA®Tesla P100 GPU — the most advanced accelerator ever built, and the first to be based on the company’s 11th generation Pascal architecture (for more on Pascal, see “Inside Pascal: NVIDIA’s Newest Computing Platform“).
- NVIDIA® While NVIDIA® is best known for our hardware platforms, their software plays a key role advancing the state of the art of GPU-accelerated computing. This body of work — the NVIDIA® SDK — has received significant updates (see “NVIDIA® Details Major Software Updates to Sharpen GPU Computing’s Cutting Edge”).
- NVIDIA® Continuing their efforts to help build autonomous vehicles with super-human levels of perception, NVIDIA® also introduced an end-to-end mapping platforms for self-driving cars. It’s designed to help automakers, map companies and start-ups rapidly create HD maps and keep them updated, using the compute power of NVIDIA® DRIVE PX 2 in the car and NVIDIA® Tesla GPUs in the data centre.