GPU servers

Optimized computing power

Take a look at our series of servers equipped with graphic cards specifically configured for resource-intensive usage. With the enhanced computing power offered by their processors, our GPU servers are designed to accelerate parallel task processing.

GPU

Advantages of GPU servers

High frequency

NVIDIA GeForce GTX 1080

By combining graphic cards and bi-processors, the server's computing power is strongly increased. With the power of thousands of combined cores, these servers are perfect for complex tasks and massively parallel processing.

Features per graphic card:

  • 8 GB DDR5
  • 2,560 CUDA cores
  • 320 Gbit/s RAM bandwidth
  • Nvidia GPU Boost 3.0 Technology
Compatible with Linux, CUDA/OpenCL and KVM.

Recommended servers

Intel  2x Xeon E5-2667v3
16c/32t - 3.2GHz /3.2GHz
64GB DDR4 ECC 2400 MHz
SoftRaid 2x480GB SSD
4 x NVIDIA Geforce GTX 1080
200 Mbps  bandwidth
vRack: 100 Mbps
From
$1 649.99
/month

Uses for a GPU server

Backup server

Cloud gaming

Create gaming environments on demand, and offer high-definition gaming sessions without worrying about server performance.

Multimedia storage

3D rendering

Harness the computing power of graphic cards to generate complex 2D and 3D animations.

High-volume database storage

Video

Accelerate processing and video encoding with the computing power of the graphic cards installed in our servers.

Your questions answered

Why use a graphic card in a server?

The processors for graphic cards (GPUs) are equipped with a much higher number of cores than a standard processor (a CPU). As a result, they can be used to perform a high number of parallel tasks. This design was initially intended for performing graphic operations (OpenGL/Direct3D). However, with programming languages like CUDA and OpenCL, it can now be used to perform tasks that would usually be processed by a CPU.

What activities can a GPU dedicated server be used for?

The architecture of a GPU server is particularly well-adapted to applications that require parallel task processing such as images, bio-computing, big data and deep learning. You can use it to drastically accelerate processing that is often very resource-intensive.