Tenants such as project, development or application support teams have access to a catalogue of images, services and virtual resources and can provision, de-provision, control (start/stop, ...) and configure all types of virtual resources such as virtual servers, disks, networks, load balancers and databases to build and manage even complex environments on their own, without involvement of system administrators. 

Resource quotas (limits), such as the maximum number of vCPUs, the maximum size of RAM or storage or the maximum IOPS can be assigned to tenants to avoid over-usage of resources and its impact on other tenants. 

Tenants may create and manage own private networks (LANs) and IP subnets, connect them to each other and to other networks with virtual routers, protect them with virtual firewalls, create virtual load balancers for high-availability and scalability of virtual servers and create their own VPN to connect securely from/to remote services.

Tenants can optionally provision ready databases (mySQL, mongo-DB, mssql, posgre) and let the cloud controllers take care of backups, patches, scaling and more. 

The status of all nodes is monitored in real-time and in case of a critical issue, an evacuation process is launched to move virtual servers and other resources with no or minimum impact on users.

In addition to the web portal, tenants may also use an extensive CLI or REST API to automate the creation, configuration, scaling and deletion of environments as part of their DevOps processes (Infrastructure-as-Code)

It is possible to measure the exact resource consumption of each tenant. This data can be used for controlling purposes (accrue costs to cost center of tenant, "bill" the tenant) or just to increase cost transparency for  infrastructure costs  (for business units, departments, project teams, ...).

Tenants can create, attach, resize, clone, snapshot, backup and restore virtual disks with different performance profiles. Disks can be also be used to store and launch virtual server images and states.

Cloud controllers are deployed in an active/active/active architecture behind a pair of load balancers. All controller services including the controller databases are clustered across three nodes.

Hardware high-availability is guaranteed through multiple measures on multiple layers:

24x7 ILO monitoring, redunant (2-4x) power supply, redunant network adapters, switches and ports using Link Layer Aggregation and redundant storage fabric elements. 

Tenants can define  environment templates and use them to deploy applications and IT systems in different scales, variants and configurations with the same template (e.g. development, QA, Performance-Test and Staging environments with the same template).

Deployment of virtual servers, disks and network components follows policy-based, automated resource scheduling processes. This relieves users from having to select the best fit from the available target hardware and leads to a well-balanced resource utilization.  

Tenants can use ready workflow services to create automation processes for and across virtual resources. 

Hardened integrations with leading container orchestration platforms and PaaS (Platform-as-a-Service) solutions are available.

The centralized cloud monitoring and management plane is connected via high availabiliy on all layers: three nodes in a cluster for each monitoring and management service, redundant switch ports and switches, redundant VPN routers and routes.