Monitoring as a Service (MaaS) gathers metrics on VM and Cube resource utilization. You can track any virtual instance in your data center and set triggers when usage limits are reached. MaaS is completely free and can be accessed immediately after a virtual instance is provisioned.

Compatibility: Monitoring is available anytime from the DCD or via API for all instances in the VDC. This service is compatible with all boot options and is independent of the operating system used or the instance type.

Actions and Alarms: With MaaS, you may set triggers based on resource usage. The system automatically executes predetermined Actions using set threshold Alarms. The application currently only sends email notifications.

Automation: MaaS lets you access up to two weeks' worth of runtime performance metrics. Raw Prometheus metrics can be gathered using TelemetryAPI (PromQL) and imported into your own monitoring system.

Collected metrics

MaaS collects basic metrics from the virtualization layer on which the virtual instance is running. The application makes it possible to run any configuration of virtual instances, as no client installation is required. Virtual instances can boot from any block storage device using public or private images as well as snapshots or boots from ISO or network. Since no client installation or specific image is required, the feature is also enabled for virtual instances that were provisioned previously. The metrics covered by MaaS are as follows:



  • Average utilization of all cores of a virtual instance


  • Ingress Network bytes/packets of all NICs of a virtual instance

  • Egress Network bytes/packets of all NICs of a virtual instance


  • Disk bytes read by the block device

  • Disk bytes written by the block device

  • Disk read IOs performed by the device

  • Disk write IOs performed by the device.

Availability of metrics

The collection of monitoring metrics starts when a new instance is provisioned. However, the first metrics will not be available immediately after provisioning. It takes about 10 minutes before the first metrics are collected and can be retrieved in DCD or via API. Once this initial 10 minute period has elapsed, you can poll for metrics at any time.


MaaS provides metric data for the two previous weeks. If you want to create a long-term history of metrics we advise you to retrieve the raw metric data and store it. Any data older than two weeks will be purged by IONOS Cloud.

IONOS currently offers CPU, Network, and Storage metrics. Memory metrics are currently not available.

Last updated