Kubernetes HPA Scale to Zero Without KEDA: Native Autoscaling for Idle Workloads
Kubernetes now supports scaling to zero replicas for idle workloads using its Horizontal Pod Autoscaler (HPA). This feature allows users to eliminate unnecessary compute costs without relying on external tools like KEDA or Knative. By setting up a few configurations, users can efficiently manage resources based on external metrics, such as queue length.
- ▪Kubernetes HPA can scale deployments to zero replicas without the need for KEDA or Knative.
- ▪To enable this feature, users need to set a feature gate and use an external metrics source.
- ▪The scaling mechanism is particularly useful for workloads that experience long idle periods, such as queue processors.
Opening excerpt (first ~120 words) tap to expand
try { if(localStorage) { let currentUser = localStorage.getItem('current_user'); if (currentUser) { currentUser = JSON.parse(currentUser); if (currentUser.id === 3330969) { document.getElementById('article-show-container').classList.add('current-user-is-article-author'); } } } } catch (e) { console.error(e); } Daniel Kraszewski Posted on May 27 • Originally published at u11d.com Kubernetes HPA Scale to Zero Without KEDA: Native Autoscaling for Idle Workloads #devops #cloud #kubernetes If you run queue processors, batch workers, or event-driven workloads that sit idle for hours between bursts, you're paying for compute you don't need. Kubernetes HPA can scale these deployments to zero replicas — no KEDA, no Knative, no external controllers required.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at DEV.to (Top).
Discussion
0 commentsMore from DEV.to (Top)
