Regions and Availability Zones
Diese Seite ist noch nicht in deiner Sprache verfügbar. Englische Seite aufrufen
What is a Region?
Section titled “What is a Region?”A region is characterized by the fact that it is completely physically separated from other regions. Each region consists of three or more availability zones, with three AZs being the rule. Furthermore, it is ensured that all AZs of a region are always located in the same country, i.e. a region is always country-specific.
The range of functions of the different products always depends on the region, so all AZs in this region have the same range of functions in each case.
Technically, it is ensured that there is at least a bandwidth of 200 GBit/s with a latency of approx. 0.5ms between the individual AZs of a region. WAN connections also terminate in one region at a time.
Currently, we offer the region “Heilbronn/Germany” which is called “eu01”.
What is an Availability Zone (AZ)?
Section titled “What is an Availability Zone (AZ)?”Each availability zone (AZ) can be assigned to exactly one region and is logically associated with that region.
An AZ is characterized by the fact that power, cooling and local network connections are each completely separated from all other AZs.
There are 3 physical conversions for AZ:
- An AZ corresponds to a data center if the above requirements are fulfilled
- An AZ is logically spanned across multiple data centers if the above requirements are fulfilled
- Several AZs can be operated in one data center if the above-mentioned requirements are fulfilled and it is also ensured that the different AZs can be operated in different fire compartments.
The above diagram shows possible characteristics of an availability zone and provides the basis for the actual availability zones.
The following diagram shows the actual layout of the EU01 region:
Availability
Section titled “Availability”Metro Availability Zone
Section titled “Metro Availability Zone”Volumes assigned to the “Metro” scope (e.g., if the VM is operated in this context) reside on a Metro storage cluster mirrored across two different data center sites.
The metro cluster is fully available via both data centers. As a result, this means that a complete AZ could be offline and the data would still be available without functional degradation.
Single Availability Zone
Section titled “Single Availability Zone”Volumes assigned to the “Single AZ” scope (e.g., if the VM is operated in this context) are located on separate storage systems within the respective selected AZ.
The storage systems in all available AZs are each fully independent of all other AZs and are equipped with high-availability controllers and SSD storage media. Accordingly, in the event of a malfunction of an entire AZ, only the respective AZ is affected by this malfunction - without any impact on the other single AZ. The storage data of a single AZ is not mirrored in another AZ - but this can be done on the application side if required.
Block Storage Regions and AZs
Section titled “Block Storage Regions and AZs”The following diagram shows the basic schematic structure of the block storage types in Region 1 (e.g. EU01).
No replication takes place between the single AZ storage systems (“Block Storage AZ”). The metro systems (“Block Storage Metro” and “Backup Storage Metro”) are each spanned across different AZs.
The Backup Storage is located on a storage system that is physically separated from the regular Block Storage.
In this way, the backups on the Backup Storage are particularly protected and independent of the primary storage systems.
