NVIDIA DGX-2
- WORLD’s FIRST 2 PETAFLOPS SYSTEM
- ENTERPRISE GRADE AI INFRASTRUCTURE
- 12 TOTAL NVSWITCHES + 8 EDR INFINIBAND/100 GbE ETHERNET
- 2 INTEL PLATINUM CPUS + 1.5 TB SYSTEM MEMORY + DUAL 10/25 GbE ETHERNET
- 30 TB NVME SSDS INTERNAL STORAGE
About
THE WORLD’S MOST POWERFUL AI SYSTEM FOR THE MOST COMPLEX AI CHALLANGES
Experience new levels of AI speed and scale with NVIDIA® DGX™-2, the first 2 petaFLOPS system that combines 16 fully interconnected GPUs for 10X the deep learning performance. It’s powered by NVIDIA® DGX™ software and a scalable architecture built on NVIDIA NVSwitch, so you can take on the world’s most complex AI challenges.
Specification
Specification
GPUs
16x Tesla V100
Performance
2 petaFLOPS
GPU Memory/Total
32 GB per GPU/512
CPU
Dual Intel Xeon Platinum
8168, 2.7 GHz, 24-cores
NVIDIA CUDA Cores
81,920
TENSOR Cores
10,240
System Memory
1.5TB
Storage
OS: 2X 960GB NVME SSDs
Internal Storage: 30TB
(8X 3.84TB) NVME SSDs
Network
8X 100Gb/sec
Infiniband/100GigE
Dual 10/25Gb/sec Ethernet
Software
Ubuntu Server Linux OS
See Software stack for details
System Weight
340 lbs
System Dimensions
795.4 D x 482.3 W x 440 H (mm)
Packing Dimensions
1180 D x 730 W x 284 H (mm)
Maximum Power Requirements
10kW
Operating Temperature Range
5 - 35 °C
You May Also Like
Related products
-
NVIDIA DGX STATION A100 320GB/160GB
SKU: DGXS-2080C+P2CMI00More Information- 2.5 petaFLOPS of performance
- World-class AI platform, with no complicated installation or IT help needed
- Server-grade, plug-and-go, and doesn’t require data center power and cooling
- 4 fully interconnected NVIDIA A100 Tensor Core GPUs and up to 320 gigabytes (GB) of GPU memory
-
NVIDIA DGX H100
SKU: DGX H100More Information- 8X NvidiaH100 Gpus With 640 Gigabytes of total gpu memory
- 4x Nvidia Nvswitches
- 8X Nvidia Connect-7 and 2x Nvidia bluefield dpu 400 gigabits-per-second network interface
- Dual x86 CPUs and 2 Terabytes of system memory
- 30 Terabytes NVME SSD
-
NVIDIA DGX Spark GB10
SKU: N/AMore Information- Built on NVIDIA GB10 Grace Blackwell Superchip
- NVIDIA Blackwell GPU with fifth-generation Tensor Core technology
- NVIDIA Grace CPU with 20- core high-performance Arm architecture
- Up to 1 petaFLOP of AI performance using FP4
- 128 GB of coherent, unified system memory
- Support for up to 200 billion parameter model
- NVIDIA ConnectX™ networking to link two systems to work with models up to 405 billion parameters
- Up to 4 TB of NVMe storage
- Compact desktop form factor
Our Customers
Previous
Next