The fastest path to Deep Learning
Building a platform for deep learning goes well beyond selecting a server and GPUs. A commitment to implementing AI in your business involves carefully selecting and integrating complex software with hardware. NVIDIA® DGX-1™ fast-tracks your initiative with a solution that works right out of the box, so you can gain insights in hours instead of weeks or months.
NVIDIA DGX-1
Explore the powerful components of DGX-1.
The first GPU architecture to incorporate Tensor Core technology designed for AI.
High-bandwidth and low-latency, with a total of 800 GB/s of communication.
For boot, storage management, and deep learning framework coordination.
High-speed interconnect 300 GB/s per GPU, 10X faster than current PCIe Gen3 x16 interconnections.
Fits in a compact rack space, using 3,200 Watts of maximum system power.
DGX-1 ARCHITECTURE ADVANCEMENTS
While many solutions offer GPU-accelerated performance, only NVIDIA DGX-1 unlocks the full potential of the latest NVIDIA GPU's like the NVIDIA Tesla® V100, including innovations like next generation NVLink and new Tensor Core architecture.
System Level Specifications
1st Gen NVIDIA DGX-1 with Tesla P100
2ND Gen NVIDIA DGX-1 with Tesla V100
TFLOPS (FP16)
170
960
CUDA Cores
28,672
40,960
Tensor Cores
--
5,120
NVLink vs PCIe Speed-up
5X
10X
Deep Learning Training Speed-up
1X
3X
ITERATE AND INNOVATE FASTER
High-performance training accelerates your productivity, which means faster time to insight and faster time to market.
EFFORTLESS PRODUCTIVITY
NVIDIA DGX-1 removes the burden of continually optimizing your deep learning software and delivers a ready-to-use, optimized software stack that can save you hundreds of thousands of pounds. It includes access to today's most popular deep learning frameworks, NVIDIA DIGITS deep learning training application, third-party accelerated solutions, the NVIDIA Deep Learning SDK (e.g. cuDNN, cuBLAS, NCCL), CUDA® toolkit, NVIDIA Docker and NVIDIA drivers.
NVIDIA DGX Software Stack
Investment Protection
With today's rapidly evolving open source software and the complexity of libraries, drivers, and hardware, it's good to know that NVIDIA's enterprise grade support and software engineering expertise are behind every DGX-1.
This software stack is built on years of R&D, innovation, and deep learning expertise, and maintained by monthly optimized framework releases. Also, NVIDIA's support includes software upgrades and priority resolution of critical issues; you can have peace of mind that your environment is tuned for maximized performance and uptime.
