High-bandwidth, energy efficient interconnect.

NVLink allows for significantly accelerated speeds and performances in your applications. With the ability to create direct connections between multiple GPUs, as well as offering flexibility in connections between CPUs and GPUs, NVLink accelerates data exchange rates ranging from 5 to 12 times faster than before with PCIe.

The higher bandwidth between GPUs offers much better multi-GPU scalability. This increased flexibility, accessibility, and efficiency send your computational power into hyperdrive, driving the future in high performance computing.

Deep Learning Devbox

Tesla P100 For SXM2

Quadro GP100

Tesla V100 For SXM2
5.3 TeraFLOPS Double Precision Performance 5.2 TeraFLOPS Double Precision Performance 7.5 TeraFLOPS Double Precision Performance
10.6 TeraFLOPS Single Precision Performance 10.3 TeraFLOPS Single Precision Performance 15 TeraFLOPS Single Precision Performance
21.2 TeraFLOPS Half Precision Performance 20.7 TeraFLOPS Half Precision Performance 120 TeraFLOPS Tensor Performance
16GB CoWos HBM2 Stacked Memory Capacity 16GB CoWos HBM2 Stacked Memory Capacity 16GB CoWos HBM2 Stacked Memory Capacity
720 GB/s CoWos HBM2 Stacked Memory Bandwidth 720 GB/s CoWos HBM2 Stacked Memory Bandwidth 900 GB/s CoWos HBM2 Stacked Memory Bandwidth
900-2H403-0000-000 VCQGP100-PB 900-2G503-0000-000
Shop Now Shop Now Shop Now

SabrePC KSR1-2000007292 NVIDIA Tesla V100 1U Rackmount Server

- (4) NVIDIA Tesla V100 SXM2 with NVLink
- (2) Intel Xeon Skylake-SP 6148 Processor
- (8) 32GB DDR4 RAM
- (2) 960GB SSD

Supermicro SYS-1029GQ-TXRT SuperServer Dual Skylake-SP 1U Rackmount Barebone System

The Supermicro SYS-1029GQ-TXRT is 1U rackmount with dual socket P support for Intel Xeon Scalable Processors.

Please contact us for customization and integration.

SabrePC CWSS-2000006551-DPN NVIDIA Quadro GP100 with 2-way NVLink Deep Learning Devbox

- (2) NVIDIA Quadro GP100 (with NVLink bridge)
- (1) Core i7-5930K 6-Core 3.5GHz 15M Processor
- (4) 16GB DDR4 memory
- (1) 256GB SSD for OS/Deep Learning Software Stacks
- (1) 4TB HDD for data
- Latest version of NVIDIA Digits, Caffe, Torch, Theano, BIDMach, CuDNNv2, OpenCV v.3.0, and NVIDIA® CUDA® Toolkit