Every Wave feature, grouped by what it works on

Continuous pod rightsizing based on actual usage

Adjusts Pod CPU/memory in place, without restart, on live workloads

AS-IS → TO-BE savings, quantified per namespace.

ML-driven real-time horizontal scaling in <10 seconds

Schedule scaling for predictable traffic events

AI-recommended min/max replica boundaries for HPA

Karpenter nodes, cost per hour, Spot ratio, and thrashing on one screen

Starts a node before the next Pod needs one, with optional image pre-pull

Keeps the first replicas on On-Demand and the rest on Spot, by replica number

Detects 11 kinds of Pod and container failures and keeps the previous log, pod.yaml, and node.yaml

Flags containers whose hourly minimum memory keeps rising, before they are OOMKilled

Records how long each Pod waited in Pending and every scheduler message behind the wait

Places each workload in one of four levels by CPU usage as a share of its request

Finds nodes that run only DaemonSet Pods, with no application Pod for about a week

Lists PersistentVolumes that no Pod uses, with capacity, reason, and estimated monthly cost

Projects Pod CPU and memory requests against node capacity, with snapshots at 7 and 30 days

Projects the usage of each PersistentVolume you choose at 7 and 30 days, next to its capacity

Sorts requests into priority tiers at the Istio gateway; lower tiers get HTTP 503 under load

Turns NetFUNNEL waiting-room segments on and off from CPU, memory, and replica thresholds

Change Wave settings from a script or from Git: a REST API and 14 Custom Resource kinds

Each engineer signs in to the Wave Console through your OIDC IdP, with a role of their own

Grows a PVC when its usage reaches your threshold, up to a max capacity you set

Deletes unused PV objects that pass every protection check, only when you turn cleanup on

Sends a Slack or HTTP webhook message when a rule you write matches

See the path from gateway to node on one map

Send Wave metrics to your existing monitoring stack