Skip to content
Google Cloud Launches Scale-to-Zero in GKE 1.37

Google Cloud Launches Scale-to-Zero in GKE 1.37

First seen 4 Oct 2026, 06:06 UTC • •
API

Article Content

Browse articles
ThreatCluster AI
ThreatCluster •October 4, 2026 at 22:01 UTC
  • •GKE 1.37 introduces native scale-to-zero capabilities for Kubernetes workloads.
  • •Workloads can scale down to zero replicas when idle and restart based on demand.
  • •Specific configurations are required for the Horizontal Pod Autoscaler to function effectively.

On October 4, 2026, Google Cloud announced the introduction of native scale-to-zero capabilities in Google Kubernetes Engine (GKE) version 1.37. This feature allows eligible workloads to scale down to zero replicas when idle and automatically restart when demand returns. The update is particularly beneficial for intermittent workloads such as batch processors and event-driven workers, which can remain idle for extended periods. The scale-to-zero feature is paired with capacity buffers to minimize startup delays while reducing idle resource consumption. However, the Horizontal Pod Autoscaler (HPA) requires specific configurations, including the use of External or Object metrics, to function correctly. Without these metrics, the autoscaler cannot effectively measure demand when no Pods are running. This update is applicable to both new and existing clusters, provided they meet the version requirements. The rollout aims to enhance resource efficiency and responsiveness in cloud environments.

Start a free Starter trial for enhanced analysis

Ask AI about this cluster

Updated just now How this analysis works

Timeline

2026-09-23
GKE 1.37 announced
Google introduced native scale-to-zero capabilities in GKE 1.37, allowing workloads to scale down to zero replicas.
Quasa
2026-10-04
GKE 1.37 officially launched
The new version includes scale-to-zero capabilities and capacity buffers to enhance resource efficiency.
www.xcademia.com

More articles in this cluster (2)

Following this threat?

Track KEDA in your own feed — you're alerted when they show up in new reporting, leak sites or exploitation.

Free account · no card needed

Extracted Entities (4)

1 / 2

Common questions

What does scale-to-zero mean?
Scale-to-zero allows Kubernetes workloads to reduce their replicas to zero when idle, conserving resources.
What configurations are needed for HPA?
The HPA requires an autoscaling/v2 configuration with minReplicas set to 0 and at least one External or Object metric.
Who benefits from this update?
Workloads that are intermittent, such as batch processors and event-driven workers, will benefit the most from this capability.