Deepfactor portal runs multiple microservices as pods in the kubernetes cluster. We provide the flexibility of updating the resource limits and hpa configuration for each of these pods. However, for most use cases, you will not need to change these values. If you are running larger workloads, please refer to the following section
Resource limits and hpa config for larger workloads
Resource limits configuration #
The following section provides the resource limits configuration for all the Deepfactor portal pods. As mentioned above, the default limits should work for most common workloads. In order to modify the resource limits, please follow the steps outlined below:
1. Create resource.yaml file
Please edit the resource limits as appropriate for your load and infrastructure.
apisvc:
initContainers:
wait:
resources:
requests:
memory: 25Mi
cpu: 25m
limits:
memory: 50Mi
cpu: 50m
containers:
resources:
requests:
memory: 250Mi
cpu: 200m
limits:
memory: 500Mi
cpu: 500m
alertstreamsvc:
initContainers:
wait:
resources:
requests:
memory: 25Mi
cpu: 25m
limits:
memory: 50Mi
cpu: 50m
containers:
resources:
requests:
memory: 500Mi
cpu: 1
limits:
memory: 1Gi
cpu: 2
authsvc:
initContainers:
wait:
resources:
requests:
memory: 25Mi
cpu: 25m
limits:
memory: 50Mi
cpu: 50m
containers:
resources:
requests:
memory: 50Mi
cpu: 25m
limits:
memory: 100Mi
cpu: 50m
eventsvc:
initContainers:
wait:
resources:
requests:
memory: 25Mi
cpu: 25m
limits:
memory: 50Mi
cpu: 50m
containers:
resources:
requests:
memory: 150Mi
cpu: 250m
limits:
memory: 250Mi
cpu: 500m
notificationsvc:
initContainers:
wait:
resources:
requests:
memory: 25Mi
cpu: 25m
limits:
memory: 50Mi
cpu: 50m
containers:
resources:
requests:
memory: 50Mi
cpu: 150m
limits:
memory: 250Mi
cpu: 500m
persistentsvc:
initContainers:
wait:
resources:
requests:
memory: 25Mi
cpu: 25m
limits:
memory: 50Mi
cpu: 50m
containers:
resources:
requests:
memory: 250Mi
cpu: 250m
limits:
memory: 500Mi
cpu: 500m
proxysvc:
initContainers:
wait:
resources:
requests:
memory: 25Mi
cpu: 25m
limits:
memory: 50Mi
cpu: 50m
containers:
resources:
requests:
memory: 100Mi
cpu: 50m
limits:
memory: 200Mi
cpu: 100m
symbolsvc:
containers:
resources:
requests:
memory: 250Mi
cpu: 250m
limits:
memory: 500Mi
cpu: 2
statussvc:
initContainers:
wait:
resources:
requests:
memory: 25Mi
cpu: 25m
limits:
memory: 50Mi
cpu: 50m
containers:
resources:
requests:
memory: 25Mi
cpu: 25m
limits:
memory: 50Mi
cpu: 50m
dfstartup:
initContainers:
wait:
resources:
requests:
memory: 25Mi
cpu: 25m
limits:
memory: 50Mi
cpu: 50m
containers:
resources:
requests:
memory: 25Mi
cpu: 25m
limits:
memory: 50Mi
cpu: 50m
policyctl:
initContainers:
wait:
resources:
requests:
memory: 25Mi
cpu: 25m
limits:
memory: 50Mi
cpu: 50m
containers:
resources:
requests:
memory: 500Mi
cpu: 250m
limits:
memory: 500Mi
cpu: 250m
ttlctl:
initContainers:
wait:
resources:
requests:
memory: 25Mi
cpu: 25m
limits:
memory: 50Mi
cpu: 50m
containers:
resources:
requests:
memory: 25Mi
cpu: 25m
limits:
memory: 50Mi
cpu: 50m
nginx:
opa:
resources:
requests:
memory: 250Mi
cpu: 250m
limits:
memory: 400Mi
cpu: 500m
containers:
resources:
requests:
memory: 250Mi
cpu: 250m
limits:
memory: 1Gi
cpu: 1
frontend:
containers:
resources:
requests:
memory: 250Mi
cpu: 250m
limits:
memory: 1Gi
cpu: 1
webappsvc:
initContainers:
wait:
resources:
requests:
memory: 25Mi
cpu: 25m
limits:
memory: 50Mi
cpu: 50m
containers:
resources:
requests:
memory: 500Mi
cpu: 250m
limits:
memory: 1Gi
cpu: 500m
archivestore:
containers:
resources:
requests:
memory: 100Mi
cpu: 50m
limits:
memory: 200Mi
cpu: 100m
clickhouse:
storage:
className: gp2
requests: 300Gi
containers:
resources:
requests:
memory: 1Gi
cpu: 1
limits:
memory: 8Gi
cpu: 4
postgres:
storage:
className: gp2
requests: 100Gi
containers:
resources:
requests:
memory: 1Gi
cpu: 1
limits:
memory: 8Gi
cpu: 4
updatersvc:
containers:
resources:
requests:
memory: 50Mi
cpu: 50m
limits:
memory: 100Mi
cpu: 100m
cert-manager:
resources:
requests:
cpu: 50m
memory: 50Mi
limits:
cpu: 100m
memory: 200Mi
webhook:
resources:
requests:
cpu: 50m
memory: 50Mi
limits:
cpu: 100m
memory: 200Mi
startupapicheck:
resources:
requests:
cpu: 25m
memory: 25Mi
limits:
cpu: 50m
memory: 100Mi
cainjector:
resources:
requests:
cpu: 50m
memory: 50Mi
limits:
cpu: 100m
memory: 200Mi
migrationsvc:
initContainers:
wait:
resources:
requests:
memory: 25Mi
cpu: 25m
limits:
memory: 50Mi
cpu: 50m
containers:
resources:
requests:
memory: 25Mi
cpu: 25m
limits:
memory: 50Mi
cpu: 50m
jobCleanup:
containers:
resources:
requests:
memory: 25Mi
cpu: 25m
limits:
memory: 50Mi
cpu: 50m
nats:
bootconfig:
resources:
requests:
memory: 25Mi
cpu: 25m
limits:
memory: 50Mi
cpu: 50m
nats:
resources:
requests:
memory: 2Gi
cpu: 2
limits:
memory: 4Gi
cpu: 4
reloader:
resources:
requests:
memory: 25Mi
cpu: 25m
limits:
memory: 50Mi
cpu: 50m
exporter:
resources:
requests:
memory: 25Mi
cpu: 25m
limits:
memory: 50Mi
cpu: 50m
natsbox:
resources:
requests:
memory: 25Mi
cpu: 25m
limits:
memory: 50Mi
cpu: 50m
2. Pass the above resource.yaml file in helm install/upgrade command
helm install df-stable deepfactor/deepfactor -n deepfactor -f override.yaml -f resource.yaml --set dfstartup.config.password=YOUR_PORTAL_PASSWORD --set dfstartup.config.portalToken= "YOUR_DEEPFACTOR_LICENSE_KEY_FROM_MY.DEEPFACTOR.IO" --set clusterName=CLUSTER_NAME_OF_YOUR_CHOICE
#
Resource limits and hpa config for larger workloads #
If you are running larger workloads, we recommend updating resource limits and hpa configuration as shown in the yaml below. We have arrived at this configuration by measuring the average and peak cpu/memory utilization for larger workloads in our labs.
hpa:
enabled: true
apisvc:
replicas: 2
alertstreamsvc:
replicas: 2
eventsvc:
replicas: 2
persistentsvc:
replicas: 2
clickhouse:
containers:
resources:
requests:
memory: 8Gi
cpu: 2
limits:
memory: 16Gi
cpu: 4
postgres:
containers:
resources:
requests:
memory: 4Gi
cpu: 1
limits:
memory: 8Gi
cpu: 4
nats:
nats:
resources:
requests:
memory: 4Gi
cpu: 2
limits:
memory: 8Gi
cpu: 4
jetstream:
memStorage:
size: 2Gi