XENON’s NITRO GK17 4GPU rack server with four Tesla P100 GPUs offers a total of 64GB of high-bandwidth GPU memory. NVLink connections between the GPUs allow for data transfers up to 80 GB/s – five times faster than PCI-Express. With the introduction of NVIDIA’s new “Pascal” Unified Memory capabilities, GPU-accelerated applications are able to access all system memory and the memory of all GPUs combined. This design empowers new applications which leverage large quantities of memory. Available fully preconfigured and preinstalled for scientific computing and Deep Learning workloads with OS, drivers, CUDA SDK, Deep Learning libraries, and NVIDIA DIGITS Deep Learning environment – ready to run GPU jobs!

Key features:

  • Up to four top-performance NVIDIA Tesla P100 GPUs with NVLink
  • NVLink connectivity for data-intensive and multi-GPU applications
  • Up to 1TB system memory
  • Available with support for high-speed interconnect fabrics such as Infiniband and Intel OmniPath (optional)
  • Supports GPUDirect RDMA


| Print

Model Number


Form Factor

Rack - 1U


Intel® C612 chipset


  • Dual socket R3 (LGA 2011) supports Intel® Xeon® processor E5-2600 v4/ v3 family
  • QPI up to 9.6GT/s

Max. Memory

  • Up to 512GB ECC 3DS LRDIMM, up to 1TB ECC RDIMM
  • 16x DIMM slots


  • 3 PCI-E 3.0 x16 slots
  • 1 PCI-E 3.0 x8 Low-profile slot


Up to 4 Tesla P100 SXM2

  • Up to 80 GB/s GPU-to-GPU NVLINK
  • Supports GPUDirect RDMA


Intel® i350 Dual port GbE LAN


  • 2x 2.5" Hot-swap drive bays
  • 2x 2.5" internal drive bays

Power Supply

2000W Redundant Power Supplies

  • Titanium Level (96%)


  • Height: 1.7" (43mm)
  • Width: 17.2" (437mm)
  • Depth: 35.2" (894mm)
  • Depth: 39.3" (997mm) with rails

Need a Quote?

Have questions about XENON’s products and solutions. Just ask. A knowledgeable Sales Specialist will get back to you shortly.

get a quote