Anfrageliste ist leer.
Produkt Filter
Produkt Typ
CPU Hersteller
CPU Onboard (e.g. Intel Atom, Intel Quark)
CPU Family
CPU Serie
Number of CPUs per node
RAM expansion
Application purpose
Enclosure height
Rack mountable enclosures
Number of nodes in one housing
Storage Drives
Maximum storage expansion
Storage Interface
Anz. interner Einschübe
Network onboard
Netzteil Features
Netzteil Leistung
Anzahl PCIe Schnittstellen
Anzahl GPUs
Gehäuse Tiefe
NVIDIA DGX Systeme
NVIDIA DGX Systems
The world’s first portfolio of purpose-built, ready-to-use deep learning systems
Inspired by the demands of deep learning and analytics, NVIDIA® DGX™ Systems are built on the new, revolutionary NVIDIA Volta™ GPU platform. Combined with innovative GPU-optimized software and simplified management, these fully integrated solutions deliver groundbreaking performance and results.
The NVIDIA DGX portfolio includes:
- DGX-2 server
Break through the barriers to AI speed and scale with NVIDIA DGX-2, the first 2 petaFLOPS system that engages 16 fully interconnected GPUs for 10X the deep learning performance. Powered by NVIDIA® DGX™ software and an architecture designed for AI-scale built on NVIDIA NVSwitch, DGX-2 enables you to take on the world’s most complex deep learning challenges. - DGX-1 server
The essential instrument for AI research, designed to accelerate your data center and streamline your deep learning workflow. Experiment faster, train larger models, and get insights starting on day one. Now offered with NVIDIA Volta™, NVIDIA DGX-1 delivers industry-leading performance for AI and deep learning. - DGX Station
NVIDIA DGX Station™ is the only personal supercomputer for leading-edge AI development. Designed for your office environment, it’s built on the same software stack powering all DGX systems, for easy experimentation, from your desk to the data center. - GPU cloud optimized software
All DGX Systems are designed to give data scientists the most powerful tools for AI exploration—tools that go from your desk to the data center to the cloud. Learn how to experiment faster, train larger models, and get insights starting on day one.
Specifications
| DGX-2 SPECIFICATIONS | DGX-1 SPECIFICATIONS | DGX STATION SPECIFICATIONS | GPUs | 16x NVIDIA® Tesla V100 | 8x Tesla V100 | 4x Tesla V100 |
|---|---|---|---|
| Performance | 2 petaFLOPS | 1 petaFLOPS | 500 terraFLOPS |
| GPU Memory | 512GB total | 256 GB total system | 128 GB total system |
| NVIDIA CUDA® Cores | 81920 | 40960 | 20480 |
| NVIDIA Tensor Cores | 10240 | 5120 | 2560 |
| CPU | Dual Intel Xeon Platinum 8168, 2.7 GHz, 24-cores | Dual 20-Core Intel Xeon E5-2698 v4 2.2 GHz | Intel Xeon E5-2698 v4 2.2 GHz (20-Core |
| Maximum Power Usage | 10 kW | 3500 W | 1500 W |
| System Memory | 1.5TB RDIMM DDR4 | 512 GB RDIMM DDR4 | 256 GB RDIMM DDR4 |
| Network | 8x 100Gb/s Infiniband 2x 10/25Gb/s Ethernet |
4x 25Gb/s Infiniband 2x 10Gb/s Ethernet |
2x 10Gb/s Ethernet |
| Storage | OS: 2x 960GB NVME SSDs Data: 30TB (8x 3.84TB) NVME SSDs |
OS/Data: 4x 1.92 TB SSD RAID 0 | OS: 1x 1.92 TB SSD Data: 3x 1.92 TB SSD RAID 0 |
| Software | Ubuntu Linux OS See Software stack for details | Ubuntu Linux OS See Software Stack for Details | Ubuntu Desktop Linux OS DGX Recommended GPU Driver CUDA Toolkit |
| System Weight | 340 lbs / 154.2 kg | 134 lbs / 60 kg | 88 lbs / 40 kg |
| System Dimensions | 796 D x 483 W x 440 H (mm) 834 D mm with Front Bezel |
866 D x 444 W x 131 H (mm) | 518 D x 256 W x 639 H (mm) |
| NVSwitches | NVLINIK via 12 NVLINK Switches | Only NVLINK | Only NVLINK |
| Acoustics | - | - | < 35 dB |
| Operating Temperature Range | 5 – 35°C | 10 – 35 °C | 10 – 30 °C |
| Display | - | - | 3x DisplayPort, 4K resolution |
| Additional Ports | - | - | 2x eSATA, 2x USB 3.1, 4x USB 3.0 |
