240620c3a4
- Scale to 0 when idle (saves money) - 15 min cooldown before scale-down - Triggers on 3+ pending HTTP requests - Targets openclaw-brain-vllm deployment
21 lines
554 B
YAML
21 lines
554 B
YAML
apiVersion: keda.sh/v1alpha1
|
|
kind: ScaledObject
|
|
metadata:
|
|
name: openclaw-brain-a100-scaling
|
|
namespace: customer1
|
|
spec:
|
|
scaleTargetRef:
|
|
name: openclaw-brain-vllm
|
|
kind: Deployment
|
|
service:
|
|
name: openclaw-brain-service
|
|
port: 8000
|
|
minReplicaCount: 0
|
|
maxReplicaCount: 1
|
|
cooldownPeriod: 900 # 15 minutes idle before scale-down
|
|
triggers:
|
|
- type: http
|
|
metadata:
|
|
path: "/v1/models"
|
|
host: "openclaw-brain-service.customer1.svc.cluster.local"
|
|
requestCount: "3" # Scale up when 3+ requests pending
|