Skip to main content

Pause and Resume a Cluster

Pause a cluster when it is not needed for a period of time. Pausing stops the cluster from processing requests and stops compute charges, while the cluster's cache capacity remains reserved and billed. Resume the cluster before sending work to it, or enable Auto-suspend to suspend an idle cluster and resume it automatically when new workload arrives.

Before you begin​

  • Before pausing a cluster, confirm that it has no active workload. A paused cluster cannot process requests.
  • A paused cluster retains its cache capacity. Cache charges continue while the cluster is paused.

Pause a cluster manually​

  1. Log in to the VeloDB Cloud console.
  2. In the upper-left corner, select the warehouse that contains the cluster.
  3. In the left navigation pane, under Compute, click Clusters.
  4. On the card of a running cluster, click Pause.

The cluster status changes to Pausing. When the operation finishes, the status changes to Paused. The cluster cannot process requests while it is paused.

Resume a cluster manually​

  1. On the Clusters page, find the paused cluster.
  2. Click Resume.
  3. In the confirmation dialog, review the restored compute resources, retained cache, and renewed charges. Click OK.

The status changes to Resuming while VeloDB Cloud restores the compute resources and mounts the retained cache. When the status changes to Running, the cluster can process requests and compute charges resume.

Configure automatic suspension​

Enable Auto-suspend when the cluster should stop after idle periods but resume as soon as new workload arrives.

  1. On the Clusters page, click the cluster card to open its details page.
  2. Click Set Auto-suspend.
  3. Enable the Auto-suspend switch.
  4. Next to Idle duration, click the Edit icon, then set the number of minutes that the cluster must remain inactive before it automatically suspends.
  5. Review the Auto Resume Trigger Condition.

With the default settings, the cluster suspends after more than 30 minutes of inactivity. The resume mode is Automatic, so the cluster resumes when new workload arrives.

Next steps​