GPU INSTANCES

Cloud servers specially designed for processing massively parallel tasks

Thumbnail

GPU instances integrate powerful graphical processors, making them an ideal choice for massively parallel processing. Since it is integrated into our Public Cloud solutions, you get the advantages of on-demand resources and hourly billing. These cloud servers are adapted to meet the most exacting requirements for machine learning and deep learning.
Combined with the integration of the Nvidia NGC platform, Tensorflow, Caffe2, Mxnet and Pytorch become simpler and quicker to use.

Thumbnail

What is an NVIDIA Tesla V100 card?

​These GPUs are some of the most powerful devices on the market, and are designed to be used in datacentres. They accelerate processing for fields such as AI and graphical processing (see details).

PERFORMANCE with NVIDIA GPU Boost

  • Double-Precision 7 teraFLOPS
  • Single-Precision 14 teraFLOPS
  • Deep Learning 112 teraFLOPS

INTERCONNECT BI-DIRECTIONAL BANDWIDTH

  • PCIe 32 GB/s

MEMORY CoWoS Stacked HBM2

  • Capacity 16 GB HBM2
  • Bandwidth 900 GB/s

GPU instances can have between one and four Nvidia Tesla V100 cards connected to them, each of which delivers guaranteed performance levels.

NVIDIA NGC is available

To provide the best user experience, OVH and NVIDIA have partnered to offer a best-in-class GPU-Accelerated platform, for deep learning and high-performance computing.​
The OVH NVIDIA GPU Cloud (NGC) combines the flexibility of the OVH Public Cloud and the NVIDIA Tesla V100 GPU. It is the simplest way to deploy and maintain GPU-accelerated containers, via a full catalogue.
NGC allows users to execute their workloads in a reliable, high-performance platform; one that respects privacy, reversibility and transparency of data location (more details)​.

What is NGC?

Thumbnail

A GPU-Accelerated Containers Catalog

Access a comprehensive catalog of GPU-accelerated containers for deep learning software, to benefit directly from the GPU instances’ power.

Thumbnail

A cloud integration based on a simple system image

Get up and running quickly and reduce the complexity typically associated with setting up software

Thumbnail

NVIDIA platform

NVIDIA deep learning containers are updated monthly, to keep systems running at peak performance. Other containers provide easy access to recent software releases.

Thumbnail

Start

Get up and running quickly and reduce the complexity typically associated with setting up software

Thumbnail

Configure

$ docker pull nvcr.io/nvidia/tensorflow
$ nvidia-docker run nvidia/tensorflow t1

Thumbnail

Run

Your AI framework is available to start processing