General Pod Autoscaler(GPA) is a extension for K8s HPA, which can be used not only for serving, also for game.

// HorizontalPodAutoscalerSpec describes the desired functionality of the HorizontalPodAutoscaler.
type HorizontalPodAutoscalerSpec struct {
	// scaleTargetRef points to the target resource to scale, and is used to the pods for which metrics
	// should be collected, as well as to actually change the replica count.
	ScaleTargetRef CrossVersionObjectReference `json:"scaleTargetRef" protobuf:"bytes,1,opt,name=scaleTargetRef"`
	// minReplicas is the lower limit for the number of replicas to which the autoscaler
	// can scale down.  It defaults to 1 pod.  minReplicas is allowed to be 0 if the
	// alpha feature gate HPAScaleToZero is enabled and at least one Object or External
	// metric is configured.  Scaling is active as long as at least one metric value is
	// available.
	// +optional
	MinReplicas *int32 `json:"minReplicas,omitempty" protobuf:"varint,2,opt,name=minReplicas"`
	// maxReplicas is the upper limit for the number of replicas to which the autoscaler can scale up.
	// It cannot be less that minReplicas.
	MaxReplicas int32 `json:"maxReplicas" protobuf:"varint,3,opt,name=maxReplicas"`
	// metrics contains the specifications for which to use to calculate the
	// desired replica count (the maximum replica count across all metrics will
	// be used).  The desired replica count is calculated multiplying the
	// ratio between the target value and the current value by the current
	// number of pods.  Ergo, metrics used must decrease as the pod count is
	// increased, and vice-versa.  See the individual metric source types for
	// more information about how each type of metric must respond.
	// If not set, the default metric will be set to 80% average CPU utilization.
	// +optional
	Metrics []MetricSpec `json:"metrics,omitempty" protobuf:"bytes,4,rep,name=metrics"`

	// behavior configures the scaling behavior of the target
	// in both Up and Down directions (scaleUp and scaleDown fields respectively).
	// If not set, the default HPAScalingRules for scale up and scale down are used.
	// +optional
	Behavior *HorizontalPodAutoscalerBehavior `json:"behavior,omitempty" protobuf:"bytes,5,opt,name=behavior"`
}

// GeneralPodAutoscalerSpec describes the desired functionality of the GeneralPodAutoscaler.
type GeneralPodAutoscalerSpec struct {
	// DrivenMode is the mode the open autoscaling mode if we do not need scaling according to metrics.
	// including MetricMode, TimeMode, EventMode, WebhookMode
	// +optional
	AutoScalingDrivenMode `json:",inline"`

	// scaleTargetRef points to the target resource to scale, and is used to the pods for which metrics
	// should be collected, as well as to actually change the replica count.
	ScaleTargetRef CrossVersionObjectReference `json:"scaleTargetRef" protobuf:"bytes,1,opt,name=scaleTargetRef"`

	// minReplicas is the lower limit for the number of replicas to which the autoscaler
	// can scale down.  It defaults to 1 pod.  minReplicas is allowed to be 0 if the
	// alpha feature gate GPAScaleToZero is enabled and at least one Object or External
	// metric is configured.  Scaling is active as long as at least one metric value is
	// available.
	// +optional
	MinReplicas *int32 `json:"minReplicas,omitempty" protobuf:"varint,2,opt,name=minReplicas"`

	// maxReplicas is the upper limit for the number of replicas to which the autoscaler can scale up.
	// It cannot be less that minReplicas.
	MaxReplicas int32 `json:"maxReplicas" protobuf:"varint,3,opt,name=maxReplicas"`

	// behavior configures the scaling behavior of the target
	// in both Up and Down directions (scaleUp and scaleDown fields respectively).
	// If not set, the default GPAScalingRules for scale up and scale down are used.
	// +optional
	Behavior *GeneralPodAutoscalerBehavior `json:"behavior,omitempty" protobuf:"bytes,4,opt,name=behavior"`
}

// ExternalAutoScalingDrivenMode defines the mode to trigger auto scaling
type AutoScalingDrivenMode struct {
	// MetricMode is the metric driven mode.
	// +optional 
	MetricMode *MetricMode `json:"metric,omitempty" protobuf:"bytes,1,opt,name=metric"`

	// Webhook defines webhook mode the allow us to revive requests to scale.
	// +optional
	WebhookMode *WebhookMode `json:"webhook,omitempty" protobuf:"bytes,2,opt,name=webhook"`

	// Time defines the time driven mode, pod would auto scale to max if time reached
	// +optional
	TimeMode *TimeMode `json:"time,omitempty" protobuf:"bytes,3,opt,name=time"`

	// EventMode is the event driven mode
	// +optional
	EventMode *EventMode `json:"event,omitempty" protobuf:"bytes,4,opt,name=event"`
}

// WebhookMode allow users to provider a server
type WebhookMode struct {
	*admregv1b.WebhookClientConfig `json:",inline"`
	// Parameters are the webhook parameters
	Parameters map[string]string `json:"parameters,omitempty" protobuf:"bytes,1,opt,name=parameters"`
}

// TimeMode is a mode allows user to define a crontab regular
type TimeMode struct {
	// TimeRanges defines a array that for time driven mode
	TimeRanges []TimeRange `json:"ranges,omitempty" protobuf:"bytes,1,opt,name=ranges"`
}

// TimeTimeRange is a mode allows user to define a crontab regular
type TimeRange struct {
// Schedule should match crontab format
Schedule string `json:"schedule,omitempty" protobuf:"bytes,1,opt,name=schedule"`

// DesiredReplicas is the desired replicas required by timemode,
DesiredReplicas int32 `json:"desiredReplicas,omitempty" protobuf:"varint,2,opt,name=desiredReplicas"`
}

// EventMode is the event driven mode
type EventMode struct {
    // Triggers are thr event triggers
    Triggers []ScaleTriggers `json:"triggers"`
}

// ScaleTriggers reference the scaler that will be used
type ScaleTriggers struct {
	// Type are the trigger type
	Type string `json:"type"`
	// Name is the trigger name
	// +optional
	Name string `json:"name,omitempty"`
	// Metadata contains the trigger config
	Metadata map[string]string `json:"metadata"`
}

// GeneralPodAutoscalerBehavior configures the scaling behavior of the target
// in both Up and Down directions (scaleUp and scaleDown fields respectively).
type GeneralPodAutoscalerBehavior struct {
	// scaleUp is scaling policy for scaling Up.
	// If not set, the default value is the higher of:
	//   * increase no more than 4 pods per 60 seconds
	//   * double the number of pods per 60 seconds
	// No stabilization is used.
	// +optional
	ScaleUp *GPAScalingRules `json:"scaleUp,omitempty" protobuf:"bytes,1,opt,name=scaleUp"`
	// scaleDown is scaling policy for scaling Down.
	// If not set, the default value is to allow to scale down to minReplicas pods, with a
	// 300 second stabilization window (i.e., the highest recommendation for
	// the last 300sec is used).
	// +optional
	ScaleDown *GPAScalingRules `json:"scaleDown,omitempty" protobuf:"bytes,2,opt,name=scaleDown"`
}

// AutoscaleRequest defines the request to webhook autoscaler endpoint
type AutoscaleRequest struct {
	// UID is used for tracing the request and response.
	UID types.UID `json:"uid"`
	// Name is the name of the workload(Squad, Statefulset...) being scaled
	Name string `json:"name"`
	// Namespace is the workload namespace
	Namespace string `json:"namespace"`
	// Parameters are the parameter that required by webhook
	Parameters map[string]string `json:"parameters"`
	// CurrentReplicas is the current replicas
	CurrentReplicas int32 `json:"currentReplicas"`
}

// AutoscaleResponse defines the response of webhook server
type AutoscaleResponse struct {
	// UID is used for tracing the request and response.
	// It should be same as it in the request.
	UID types.UID `json:"uid"`
	// Set to false if should not do scaling
	Scale bool `json:"scale"`
	// Replicas is targeted replica count from the webhookServer
	Replicas int32 `json:"replicas"`
}

// AutoscaleReview is passed to the webhook with a populated Request value,
// and then returned with a populated Response.
type AutoscaleReview struct {
	Request  *AutoscaleRequest  `json:"request"`
	Response *AutoscaleResponse `json:"response"`
}