Virtual Machine Instances

OVHcloud offers a wide range of instances to meet your computing, memory, storage, and performance needs.

• General Purpose instances – these instances provide a balance between computing power, memory, and storage. They are suitable for a wide range of applications and workloads.
   
• Compute Optimized instances – these instances are designed for compute-intensive workloads, such as data processing, modeling, simulation, and rendering.
   
• Memory Optimized Instances – these instances are designed for memory-intensive applications, such as in-memory databases and data analytics. They offer a high amount of RAM.
   
• Graphics Processing Unit (GPU) instances – these instances are equipped with graphics processing units (GPUs), making them ideal for machine learning, 3D visualization, modeling, and simulation.

While Discovery instances have a 99.95% SLA, other instance types offer a monthly availability of 99.99%. For further information, please refer to the General Terms and Conditions of Service.

Follow our step-by-step guide to seamlessly create a cloud instance. Read guide

For 3rd generation instances compatible with Windows Server (B3, C3, R3), Windows Server license billing is separate from the instance.

On your invoice, you will therefore see two separate lines:

1. A line for your instance consumption.
2. A line for Windows license consumption calculated per vCPU.

Example: For a B3-16 instance with Windows Server (4 vCores and 16 GB of RAM):

• You will be billed based on the hourly cost of the B3-16 instance.
• Windows license will be billed hourly for 4 vCore.

For better readability, the total price per hour, including instance and license is displayed directly on our Pricing page.

GPU instances are virtualized by the KVM hypervisor in the Linux kernel. KVM is an open-source solution that has been proven reliable for many years. VMs are fully virtualized and do not require any guest system changes. KVM also offers advanced features such as live migration for virtual machines.

Yes. You can increase the size of Discovery instances. However, the size cannot be reduced.

No. Discovery instances are used the same way as other instances and offer the same features.

All of our compute instances are available in Vint Hill, VA, and Hillsboro, OR. General Purpose compute instances are available in Vint Hill, Hillsboro, and all US Local Zones.

Yes. You can use the OpenStack API “shelve” function to automatically back up your instance, save the configuration, and free up resources. When you use the “unshelve” function, the same process will be carried out in reverse to restore the instance to its original state.

Choosing the right cloud instance will depend on the balance of budget, performance, and needs for your application. We typically recommend starting with an iterative approach, monitoring application performance closely, and tuning instances as requirements change. Want to try our instances? Create your first free Public Cloud project! Try out Public Cloud.

A server instance is a virtual server unit allocated within a cloud computing environment. In other words, it is a virtual server that functions as if it were a physical server but is actually hosted on a shared physical server infrastructure. Server instances are created using virtualization, which allows multiple instances to coexist on a single physical server without interference.

A virtual machine (VM) is a computing resource that mimics a physical system and runs on a software called a hypervisor. Hypervisors are used to run several VMs on a single physical server. Each VM has its own operating system, applications, and isolated resources (CPU, memory, and storage). In addition to offering flexibility, isolation, and efficiency, VMs make deploying applications simple, and testing and managing IT environments intuitive.