Extensive computing power is required in order to be able to take advantage of AI, and one system that can meet this challenge is the NVIDIA® DGX™, based on the high performance NVIDIA Volta™ GPU platform. The NVIDIA DGX offers companies GPU-optimised software and simplified management all in one compact system.
Based on the latest NVIDIA Volta™ GPU architecture
Deep learning training, inference and acceleration of analyses in one single system
Unmatched performance for faster iterations and innovations
Which GPUs offer the best performance?
There are two processors that are particularly suitable for AI: NVIDIA® Tesla® V100 is the most advanced data centre GPU ever built to accelerate AI, HPC, and graphics. Best of all: the new Volta™ GV100 GPU is already integrated meaning even more performance. GPUs with Volta architecture come with over 640 Tensor Cores and have 100 teraFLOPS (TFLOPS) of deep learning performance—more than five times that of its predecessor, Pascal.
The Tesla® P40, on the other hand, is designed for maximum throughput for ever faster growing data volumes such as big data.
The NVIDIA® Tesla® V100 is also particularly recommended for High Performance Computing (HPC), whereas the NVIDIA® Tesla® P100 is ideal to accelerate HPC and AI.
For graphics virtualisation, the NVIDIA® Tesla P6 is designed specifically for blade servers and supports multiple data centre loads and also deep learning and HPC. The NVIDIA® Tesla® P4taps into the industry-leading NVIDIA Pascal™ architecture to deliver up to twice the professional graphics performance of the NVIDIA® Tesla® M60.
Tried and tested:
Some of the world's biggest data centres are already using Tesla GPU to improve performance. The Tesla platform also supports standard applications and system management tools, making it easier for CIOs to ensure maximum system availability and performance.