diff --git a/api/v1alpha1/variantautoscaling_types.go b/api/v1alpha1/variantautoscaling_types.go index cb699e1e4..0c2031bf9 100644 --- a/api/v1alpha1/variantautoscaling_types.go +++ b/api/v1alpha1/variantautoscaling_types.go @@ -72,8 +72,9 @@ type OptimizedAlloc struct { LastRunTime metav1.Time `json:"lastRunTime,omitempty"` // Accelerator is the type of accelerator for the optimized allocation. - // +kubebuilder:validation:MinLength=2 - Accelerator string `json:"accelerator"` + // Deprecated: This field is deprecated and will be removed in a future version. Use node selector or node affinity from scale target instead. + // +optional + Accelerator string `json:"accelerator,omitempty"` // NumReplicas is the number of replicas for the optimized allocation. // nil means no optimization decision has been made yet. diff --git a/charts/workload-variant-autoscaler/crds/llmd.ai_variantautoscalings.yaml b/charts/workload-variant-autoscaler/crds/llmd.ai_variantautoscalings.yaml index b3b4ad5b4..5f92a5de8 100644 --- a/charts/workload-variant-autoscaler/crds/llmd.ai_variantautoscalings.yaml +++ b/charts/workload-variant-autoscaler/crds/llmd.ai_variantautoscalings.yaml @@ -201,9 +201,9 @@ spec: allocation based on autoscaling logic. properties: accelerator: - description: Accelerator is the type of accelerator for the optimized - allocation. - minLength: 2 + description: |- + Accelerator is the type of accelerator for the optimized allocation. + Deprecated: This field is deprecated and will be removed in a future version. Use node selector or node affinity from scale target instead. type: string lastRunTime: description: LastRunTime is the timestamp of the last optimization @@ -217,8 +217,6 @@ spec: format: int32 minimum: 0 type: integer - required: - - accelerator type: object type: object type: object diff --git a/config/crd/bases/llmd.ai_variantautoscalings.yaml b/config/crd/bases/llmd.ai_variantautoscalings.yaml index b3b4ad5b4..5f92a5de8 100644 --- a/config/crd/bases/llmd.ai_variantautoscalings.yaml +++ b/config/crd/bases/llmd.ai_variantautoscalings.yaml @@ -201,9 +201,9 @@ spec: allocation based on autoscaling logic. properties: accelerator: - description: Accelerator is the type of accelerator for the optimized - allocation. - minLength: 2 + description: |- + Accelerator is the type of accelerator for the optimized allocation. + Deprecated: This field is deprecated and will be removed in a future version. Use node selector or node affinity from scale target instead. type: string lastRunTime: description: LastRunTime is the timestamp of the last optimization @@ -217,8 +217,6 @@ spec: format: int32 minimum: 0 type: integer - required: - - accelerator type: object type: object type: object diff --git a/docs/user-guide/crd-reference.md b/docs/user-guide/crd-reference.md index d0d24fcf4..b32a59ded 100644 --- a/docs/user-guide/crd-reference.md +++ b/docs/user-guide/crd-reference.md @@ -44,7 +44,7 @@ _Appears in:_ | Field | Description | Default | Validation | | --- | --- | --- | --- | | `lastRunTime` _[Time](https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.32/#time-v1-meta)_ | LastRunTime is the timestamp of the last optimization run. | | | -| `accelerator` _string_ | Accelerator is the type of accelerator for the optimized allocation. | | MinLength: 2
| +| `accelerator` _string_ | Accelerator is the type of accelerator for the optimized allocation. This field is deprecated and will be removed in a future version. Use node selector or node affinity from scale target instead. | | | | `numReplicas` _integer_ | NumReplicas is the number of replicas for the optimized allocation.
nil means no optimization decision has been made yet. | | Minimum: 0
|