Cross-Namespace Grouping via Labels

PerfectScale now supports cross-namespace grouping for ephemeral workloads. You can group workloads from multiple namespaces under a single target namespace using labels and manage automation for all of them in one place.

Why this matters

In environments with dynamic namespace structures, applying consistent automation policies across related workloads is difficult. Short-lived workloads from tools like Spark, GitLab, and Airflow often span multiple namespaces, making centralized management impractical — until now.

How it works

1. Create a target namespace as the grouping destination.
2. Apply grouping labels to relevant workloads, including perfectscale.io/workload-grouping-workload-namespace with the target namespace value.
3. Define namespace-level or workload-level automation configuration in the target namespace.
4. PerfectScale groups the labeled workloads under the target namespace and applies the configured automation.

Example pod configuration

apiVersion: v1
kind: Pod
metadata:
  name: pod-name
  namespace: {original-namespace}
  labels:
    perfectscale.io/workload-grouping-workload-name: {target-name}
    perfectscale.io/workload-grouping-workload-type: {target-type}
    perfectscale.io/workload-grouping-workload-namespace: {target-namespace}

Key details

• The workload-grouping-workload-namespace label is optional. Without it, workloads remain grouped within their original namespace.
• When the target namespace label is set, automation configs from the original namespace are ignored. Only configs from the target namespace apply.
• Automation will only reduce resources after PerfectScale has collected sufficient data.
• This feature requires autoscaler version 1.0.43 or later.

Read the full documentation on cross-namespace grouping.