Skip to content

fact.instance_metering

The fact.instance_metering contains all the information for instance level metering. The data is stored as intervals where events are valid (examples given below). The intervals get updated on every metering run to the latest load date.

A few examples of valid events for any cluster:

  1. Created > Resumed > Suspended
  2. Created > Resumed > Dropped
  3. Altered > Resumed > Suspended
  4. Altered > Resumed > Dropped
Column NameData TypeDescriptionExample
metering_runTIMESTAMPTZAssigned on each load of the table from the curr_load_ts.2023-12-21T05:50:12.650643Z
datetime_startTIMESTAMPTZThe date time when the cluster started. One row is created for each cluster even if the event crosses multiple days.2023-12-21T05:31:47Z
datetime_endTIMESTAMPTZThe end time when the cluster is dropped/ suspended/ altered.2023-12-21T05:50:12Z
user_skBIGINT NOT NULLA sequencer generated unique ID for each user. Foreign key to dim.user21504
cluster_skBIGINT NOT NULLA sequencer generated unique ID for each cluster. Foreign key to dim.cluster pointing to the actual config of the cluster.25600
hardware_instance_skBIGINT NOT NULLA sequencer generated unique ID for each hardware instance. Foreign key to dim.hardware_instance tracking which hardware instance, the cluster is running on.7168
vcpu_coresBIGINT NOT NULLThe number of vCPU provided for each node of the cluster based on hardware instance type.16
node_countBIGINTNumber of nodes for the cluster as requested by the user. The user can change the number of nodes by altering the cluster.1
vcpu_secondsBIGINTTotal number of vcpu_seconds consumed by the cluster.17680