public cloud data ovh

GPU-accelerated containers for AI and HPC

OVHcloud and NVIDIA have partnered to deliver the best GPU acceleration platform, optimised for deep learning and high-performance computing.

OVHcloud's NVIDIA GPU Cloud (NGC) combines the flexibility of the Public Cloud with the power of the NVIDIA Tesla V100 graphics card, providing a complete catalog of GPU-accelerated containers that can be deployed and maintained for artificial intelligence applications.

It enables users to run their projects on a reliable and efficient platform that respects confidentiality, reversibility and transparency.

GPU Accelerated Containers

Get a complete catalogue of GPU-accelerated containers for deep-learning software, applications and HPC visualization.

 

Innovate in a few minutes

Get your projects up and running quickly and easily without the complexity typically associated with software configuration.

 

Stay up-to-date

The deep-learning NVIDIA containers are updated monthly for optimal operation. Others allow you to easily access recent software versions.

 

Icons/concept/Page/Page Certificate Created with Sketch.

ISO/IEC 27001, 27701 and health data hosting compliance*

Our cloud infrastructures and services are ISO/IEC 27001, 27017, 27018 and 27701 certified. Thanks to our compliance, you can host healthcare data securely.

* Coming soon

What is the OVHcloud NVIDIA GPU Cloud (NGC)?

NGC Arbo

Our NVIDIA GPU Cloud (NGC) provides researchers and data scientists with easy access to a comprehensive catalogue of GPU software tools optimised for deep learning and high-performance computing (HPC), which take full advantage of NVIDIA hardware. The NGC Container Registry includes NVIDIA containers optimised, tested, certified and maintained for the most popular deep learning frameworks. It also offers third-party managed HPC application containers, NVIDIA HPC visualisation containers and partner applications.

How to

1

Start

Launch your instance by selecting your preferred T1 model and NGC image.

2

Configure

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

3

Use

Your AI framework is ready to go!

Ready to get started?

Create an account and launch your services in minutes

Features

High performance, high availability

Guaranteed resources 100% of the time, with no over-allocation.

Simplicity

Do you need a single server or a complete infrastructure? The OVH Control Panel lets you choose. Billing is simple and transparent, with hourly or monthly options for each server.

Flexibility

Adapt your instances to your needs by adding disks and IP addresses, or moving them from one instance to another, whenever you need to.

Reversibility

OVHcloud uses, contributes to and supports OpenStack APIs, which ensure that your cloud environments can be effortlessly migrated, and securely connected to solutions from other cloud providers.

Pricing Public Cloud

NVIDIA NGC Platform pricing

NGC is available on certain GPU instances, which are invoiced as normal, although NGC is free.

Your questions answered

What SLA does OVHcloud offer for accessing the NVIDIA NGC Platform service?

Although we maintain a high quality of service, this solution is not managed, and OVHcloud does not guarantee availability. For further information, please refer to the Terms & conditions.

What guarantees are offered for the resources (compute, storage, etc.) used by the service?

This service is built on other cloud resources that have their own SLAs, and can be accessed from their respective pages.

Are there licence costs added to the applications offered?

No. NVIDIA only offers free applications, and they are usually open-source.

Where can I find the list of available applications?

NVIDIA offers dozens of solutions for machine learning and deep learning. Their online catalogue offers everything you can use with NGC. To get a list of packaged applications only, select the container filter: https://ngc.nvidia.com/catalog/

Are the applications maintained and managed?

No. Application images are regularly updated, but once they are deployed, they are not maintained. You can use a newer version of the image to get the latest updates for your application.