Skip to content

Concepts of Block Storage

Use, create and mount volumes/datastores (Block Storage - Root & Data Volumes)

Section titled “Use, create and mount volumes/datastores (Block Storage - Root & Data Volumes)”

The first variant of the block storage comprises the operating system memory (“image”), which is already selected by the choice of flavor. This storage cannot be subsequently modified or deleted - only deletion of the entire virtual machine (including block storage) and possible creation of a new one is possible.

The second variant of block storage is technically the same type of storage (block storage), but this storage can be optionally added or removed at any time. This additional storage can be configured as needed with sizes ranging from 1GB to 10TB per volume. The minimum change rate is 1GB, although an increase is possible at any time as long as the volume is not attached to any instance. Reduction of existing volume capacity is not possible. However, an additional, smaller memory can be added at any time, allowing the larger memory to be decoupled (and deleted if necessary) after a data copy. This form of storage can only be attached to a single virtual machine at a time.

Snapshots, full backups and incremental backups

Section titled “Snapshots, full backups and incremental backups”

Snapshots, full backups and incremental backups can be created for volumes. The relevant details can be viewed here: Introduction to STACKIT Server

Block storage for operating system images (“Images”)

Section titled “Block storage for operating system images (“Images”)”

On the customer side, images can also be uploaded to the STACKIT Cloud and used as the basis for new VMs. There are no guaranteed performance classes for this image storage.

For uploading and using this function, access to and use of the API is necessary (see IaaS API)

A region represents a physical data center or multiple data centers that are located in geographical proximity to each other. Each region is completely independent to improve fault tolerance and stability.

After a resource is created, its region cannot be changed. An Availability Zone is defined as a physical location that is using independent power supplies and networks. If any hardware or software faults occur in one AZ, it does not affect other AZs. Each region consists of at least three Availability Zones (AZs).

Connect and disconnect data volumes/datastores with VMs

Section titled “Connect and disconnect data volumes/datastores with VMs”

Data volumes can be attached to VMs or detached. In both cases, the system should be shut down in advance. It is also possible to detach a data volume from a VM and connect it to another VM. You can find detailed instructions about the function here: Attach an additional disk

It is not possible to connect and disconnect root volumes because these volumes are permanently connected to the VM.

System-local data carriers (ephemeral disk)

Section titled “System-local data carriers (ephemeral disk)”

Using the infrastructure API, it is possible to create and use data carriers locally on the hypervisor. In this case, the data is only stored singularly (not redundant, not mirrored, not multiple, not encrypted); protection against failure or loss of this data is not provided by STACKIT.