Introduction
Automatically expand Kubernetes persistent volume capacity and rightsize IOPS and throughput with Cast AI Storage Autoscaling.
Early Access FeatureThis feature is in early access. It may undergo changes based on user feedback and continued development. We recommend testing in non-production environments first and welcome your feedback to help us improve.
Overview
Cast AI Storage Autoscaling automatically expands persistent volume capacity and rightsizes IOPS and throughput for Kubernetes workloads on GKE and EKS. It reduces storage costs by eliminating over-provisioned resources while ensuring that applications have the disk performance they need.
Storage Autoscaling runs in dry-run mode by default, so you can review planned actions before enabling live changes. For details on how capacity expansion and performance rightsizing work, see How it works.
Key benefits
- Automatic capacity expansion. Grows PVs before they run out of space, based on configurable usage thresholds.
- IOPS and throughput rightsizing. Aligns provisioned performance with actual workload demand on supported disk types.
- Cloud-native support. Works with GKE Hyperdisk and Persistent Disk volumes, and with EBS gp3, io1, and io2 volumes on EKS.
- Dry-run safety mode. Installs without making changes so you can review planned actions before enabling live mode.
- Policy-based configuration. Define optimization rules using PVPolicy Custom Resources to control which PVCs are managed and how.
Supported cloud providers and disk types
Capacity expansion is available for all listed types. Independent IOPS and throughput tuning is available only on the high-performance types.
GKE
- Hyperdisk Balanced, Extreme, and Throughput
- Persistent Disk Extreme, Balanced, SSD, and Standard
EKS
- gp3, io1, and io2 (capacity, IOPS, and throughput tuning)
Get started
Updated 4 hours ago