danswer/deployment/cloud_kubernetes/keda/indexing-model-server-scaledobject.yaml
pablodanswer facf1d55a0
Cloud improvements (#3099)
* add improved cloud configuration

* fix typing

* finalize slackbot improvements

* minor update

* finalized keda

* moderate slackbot switch

* update some configs

* revert

* include reset engine!
2024-11-13 23:52:52 +00:00

20 lines
469 B
YAML

apiVersion: keda.sh/v1alpha1
kind: ScaledObject
metadata:
name: indexing-model-server-scaledobject
namespace: danswer
labels:
app: indexing-model-server
spec:
scaleTargetRef:
name: indexing-model-server-deployment
pollingInterval: 15 # Check every 15 seconds
cooldownPeriod: 30 # Wait 30 seconds before scaling down
minReplicaCount: 1
maxReplicaCount: 14
triggers:
- type: cpu
metadata:
type: Utilization
value: "70"