Quick tech specs
- Deep Learning
- Performance: 2 petaFLOPS
- NVIDIA Tensor Cores: 10240
- Processor: Dual Intel Xeon Platinum 8168
- GPUs: 16x NVIDIA® Tesla V100
- NVIDIA CUDA® Cores: 81920
- NVSwitches: 12
- Network: 8X 100Gbps Infiniband/100GigE Dual 10/25Gbps E
Know your gear
The NVIDIA® DGX-2 16x Deep Learning Server is specifically engineered for maximum performance for demanding applications. It is a 2 petaFLOPS system that combines 16 interconnected GPUs for high levels of speed and scale. Powered by NVIDIA DGX? software and the scalable architecture of NVIDIA, it is the platform of choice for taking on complex AI challenges. This server is purpose-built for reliability, availability and serviceability (RAS) to reduce unplanned downtime, streamline serviceability and maintain operational continuity.