Nvidia unveils DGX-1 - The World's First Deep Learning Supercomputer in a Box

http://www.nvidia.com/object/deep-learning-system.html

GPUs : 8x Tesla GP100
TFLOPS (GPU FP16 / CPU FP32) : 170/3
GPU Memory : 16 GB per GPU
CPU : Dual 16-core Intel Xeon E5-2698 v3 2.3 GHz
NVIDIA CUDA® Cores : 28672
System Memory : 512 GB 2133 MHz DDR4 LRDIMM
Storage : 4x 1.92 TB SSD RAID 0
Network : Dual 10 GbE, 4 IB EDR
Software : Ubuntu Server Linux OS, DGX-1 Recommended GPU Driver
System Weight : 134 lbs
System Dimensions : 866 D x 444 W x 131 H (mm)
Packing Dimensions : 1180 D x 730 W x 284 H (mm)
Maximum Power Requirements : 3200W
Operating Temperature Range : 10 - 35 °C

Note : 134 pounds is 60kg.


3200W power requirement ... *whistles*.
My Seasonic PSU definitely won't be able to support this 

There might be potential applications for this in GPU mining, depending on the price of this hardware.