Skip to content


Besides single machines, SimpleVM users can start entire clusters. Since the cluster feature is still in the development phase, only selected projects can use it. Contact if you want the cluster feature activated for your project.

The cluster feature enables the orchestration of several virtual machines to distribute tasks to instances. It distinguishes between a master instance and worker nodes. The flavor of each node and the flavor of the master instance can differ. Only the image is the same for all instances. The instances are connected to each other and can thus exchange data.

In SimpleVM, you can add worker nodes to your cluster, called up scaling. You can remove worker nodes from your cluster, called down scaling. See the instructions here.

Shared directories

The master node provides shared spool-disk space between the master and all worker nodes. You can find it under the path /vol/spool.
/vol/scratch/ corresponds to the path of the ephemeral disk, thus the local disk space of a single worker node or a master node.

Start and manage your clusters

In SimpleVM, you use clusters similarly to your usual virtual machines. Start a cluster on the "New Cluster" page and manage your cluster on the "Cluster overview". See New Cluster for information on starting a cluster, see Cluster Overview for information about the cluster overview.

OpenStack and BiBiGrid

If you belong to an OpenStack project, see this BiBiGrid tutorial on how to use BiBiGrid. SimpleVM uses BiBiGrid to offer the cluster feature.

Last update: December 28, 2022