@@ -66,161 +66,6 @@ spec:
6666 description : Spec defines the desired state for autoscaling the model
6767 variant.
6868 properties :
69- behavior :
70- description : |-
71- Behavior configures the HPA scaling behavior policies (scale-up and scale-down).
72- When omitted, default Kubernetes HPA scaling behavior is used.
73- Applied directly to HPA, or passed through to KEDA's underlying HPA via
74- spec.advanced.horizontalPodAutoscalerConfig.behavior.
75- properties :
76- scaleDown :
77- description : |-
78- scaleDown is scaling policy for scaling Down.
79- If not set, the default value is to allow to scale down to minReplicas pods, with a
80- 300 second stabilization window (i.e., the highest recommendation for
81- the last 300sec is used).
82- properties :
83- policies :
84- description : |-
85- policies is a list of potential scaling polices which can be used during scaling.
86- If not set, use the default values:
87- - For scale up: allow doubling the number of pods, or an absolute change of 4 pods in a 15s window.
88- - For scale down: allow all pods to be removed in a 15s window.
89- items :
90- description : HPAScalingPolicy is a single policy which must
91- hold true for a specified past interval.
92- properties :
93- periodSeconds :
94- description : |-
95- periodSeconds specifies the window of time for which the policy should hold true.
96- PeriodSeconds must be greater than zero and less than or equal to 1800 (30 min).
97- format : int32
98- type : integer
99- type :
100- description : type is used to specify the scaling policy.
101- type : string
102- value :
103- description : |-
104- value contains the amount of change which is permitted by the policy.
105- It must be greater than zero
106- format : int32
107- type : integer
108- required :
109- - periodSeconds
110- - type
111- - value
112- type : object
113- type : array
114- x-kubernetes-list-type : atomic
115- selectPolicy :
116- description : |-
117- selectPolicy is used to specify which policy should be used.
118- If not set, the default value Max is used.
119- type : string
120- stabilizationWindowSeconds :
121- description : |-
122- stabilizationWindowSeconds is the number of seconds for which past recommendations should be
123- considered while scaling up or scaling down.
124- StabilizationWindowSeconds must be greater than or equal to zero and less than or equal to 3600 (one hour).
125- If not set, use the default values:
126- - For scale up: 0 (i.e. no stabilization is done).
127- - For scale down: 300 (i.e. the stabilization window is 300 seconds long).
128- format : int32
129- type : integer
130- tolerance :
131- anyOf :
132- - type : integer
133- - type : string
134- description : |-
135- tolerance is the tolerance on the ratio between the current and desired
136- metric value under which no updates are made to the desired number of
137- replicas (e.g. 0.01 for 1%). Must be greater than or equal to zero. If not
138- set, the default cluster-wide tolerance is applied (by default 10%).
139-
140- For example, if autoscaling is configured with a memory consumption target of 100Mi,
141- and scale-down and scale-up tolerances of 5% and 1% respectively, scaling will be
142- triggered when the actual consumption falls below 95Mi or exceeds 101Mi.
143-
144- This is an alpha field and requires enabling the HPAConfigurableTolerance
145- feature gate.
146- pattern : ^(\+|-)?(([0-9]+(\.[0-9]*)?)|(\.[0-9]+))(([KMGTPE]i)|[numkMGTPE]|([eE](\+|-)?(([0-9]+(\.[0-9]*)?)|(\.[0-9]+))))?$
147- x-kubernetes-int-or-string : true
148- type : object
149- scaleUp :
150- description : |-
151- scaleUp is scaling policy for scaling Up.
152- If not set, the default value is the higher of:
153- * increase no more than 4 pods per 60 seconds
154- * double the number of pods per 60 seconds
155- No stabilization is used.
156- properties :
157- policies :
158- description : |-
159- policies is a list of potential scaling polices which can be used during scaling.
160- If not set, use the default values:
161- - For scale up: allow doubling the number of pods, or an absolute change of 4 pods in a 15s window.
162- - For scale down: allow all pods to be removed in a 15s window.
163- items :
164- description : HPAScalingPolicy is a single policy which must
165- hold true for a specified past interval.
166- properties :
167- periodSeconds :
168- description : |-
169- periodSeconds specifies the window of time for which the policy should hold true.
170- PeriodSeconds must be greater than zero and less than or equal to 1800 (30 min).
171- format : int32
172- type : integer
173- type :
174- description : type is used to specify the scaling policy.
175- type : string
176- value :
177- description : |-
178- value contains the amount of change which is permitted by the policy.
179- It must be greater than zero
180- format : int32
181- type : integer
182- required :
183- - periodSeconds
184- - type
185- - value
186- type : object
187- type : array
188- x-kubernetes-list-type : atomic
189- selectPolicy :
190- description : |-
191- selectPolicy is used to specify which policy should be used.
192- If not set, the default value Max is used.
193- type : string
194- stabilizationWindowSeconds :
195- description : |-
196- stabilizationWindowSeconds is the number of seconds for which past recommendations should be
197- considered while scaling up or scaling down.
198- StabilizationWindowSeconds must be greater than or equal to zero and less than or equal to 3600 (one hour).
199- If not set, use the default values:
200- - For scale up: 0 (i.e. no stabilization is done).
201- - For scale down: 300 (i.e. the stabilization window is 300 seconds long).
202- format : int32
203- type : integer
204- tolerance :
205- anyOf :
206- - type : integer
207- - type : string
208- description : |-
209- tolerance is the tolerance on the ratio between the current and desired
210- metric value under which no updates are made to the desired number of
211- replicas (e.g. 0.01 for 1%). Must be greater than or equal to zero. If not
212- set, the default cluster-wide tolerance is applied (by default 10%).
213-
214- For example, if autoscaling is configured with a memory consumption target of 100Mi,
215- and scale-down and scale-up tolerances of 5% and 1% respectively, scaling will be
216- triggered when the actual consumption falls below 95Mi or exceeds 101Mi.
217-
218- This is an alpha field and requires enabling the HPAConfigurableTolerance
219- feature gate.
220- pattern : ^(\+|-)?(([0-9]+(\.[0-9]*)?)|(\.[0-9]+))(([KMGTPE]i)|[numkMGTPE]|([eE](\+|-)?(([0-9]+(\.[0-9]*)?)|(\.[0-9]+))))?$
221- x-kubernetes-int-or-string : true
222- type : object
223- type : object
22469 maxReplicas :
22570 default : 2
22671 description : |-
0 commit comments