The Kubernetes autoscaler default mode is to attempt to consolidate workloads to run the same applications on fewer nodes. However, when nodes contain unevictable Pods—Pods that cannot be safely evicted due to constraints such as Pod Disruption Budgets or eviction restrictions (for example, safe-to-evict settings)—those nodes cannot be drained.
As a result, Kubernetes clusters often suffer from Pod fragmentation, where Pods are spread inefficiently across nodes and unused capacity is distributed instead of consolidated.
These Pods become effectively pinned to nodes, blocking autoscaler scale-down.
Pod fragmentation directly limits the effectiveness of Pod rightsizing. Even when workloads themselves are fully optimized, inefficient Pod placement can prevent node consolidation, which leads to waste in the cluster.
You can see the average rate of waste in the Dashboard and Clusters > Overview tabs.
Don’t see the average waste?
To see the average waste figure, the installed Insights agent must be v0.164.1 or later. In earlier versions, N/A is displayed.
Adaptive Pod placement addresses this by improving how Pods are distributed across the cluster. It classifies Pods as either evictable or unevictable, and then adjusts their placement by grouping similar Pods and separating Pods of different classifications to reduce fragmentation and improve packing efficiency.
By enabling APP, you can:
Reduce fragmentation that blocks node scale-down.
Improve Pod packing across nodes.
Increase the effectiveness of the cluster autoscaler.
Enable more consistent and predictable workload placement.
You can see the potential savings from using APP in the Zesty platform Clusters > Workloads tab or by running the Status check.
Adaptive Pod placement benefits require node consolidation to be enabled. Workloads scheduled only onto node pools or ASGs without consolidation will not realize these benefits.
See also: