Scale to Zero Replicas

Overview

Scale to zero stops all workloads except the operator pod. This reduces costs but makes the cluster unavailable until scaled back up.

When scaled to zero:

• No API or UI access
• Queries and ingest stop
• Good for cost savings during downtime, not for production hours

⚠️

Scale to zero is not recommended for use in production environments.

Scale to zero replicas

To autoscale everything in a cluster off except for the operator pod, add this line to the top level spec:

scale_off: true

Only the operator pod remains running, to allow scaling back up.

spec:
  kubernetes_profile: gcp
  hydrolix_url: https://host.hydrolix.net
  env:
  db_bucket_region: us-central1
  scale_off: true   # Turn everything off

It takes a few minutes for all components to scale down.

To scale back up, remove scale_off: true and apply the changes to the hydrolixcluster.yaml file.