Version: Latest
Resource Requirements
Below are the recommended CPU and Memory requirements for the individual pods. You can experiment with different values according to your cluster usage metrics and adjust accordingly. These values can be set under the resources section under each component in the values.yaml file
resources:
limits:
cpu: 500m
memory: 500Mi
requests:
cpu: 500m
memory: 500Mi
| Pod | CPU | Memory |
|---|---|---|
| backend | 1000m | 1000Mi |
| keycloak | 1000m | 1000Mi |
| web-client | minimal/default | minimal/default |
| event-ingestion | minimal/default | minimal/default |
| nginx-reverse-proxy | minimal/default | minimal/default |
| model-training-orchestrator | minimal/default | minimal/default |
| model-training-consumer | minimal/default | minimal/default |
| model-running-orchestrator | minimal/default | minimal/default |
| model-running-orchestrator | minimal/default | minimal/default |
Resources for the Rasa Pro used for training and running your assistant model need to be specified in separate environment variables called RASA_REQUESTS_CPU RASA_REQUESTS_MEMORY RASA_LIMITS_CPU and RASA_LIMITS_MEMORY
They are allocated 1000m CPU and 1Gi memory by default. We recommend increasing this value if you are training larger models with NLU data (NLU Triggers) since the training process can be resource-intensive.