BentoDeployment CRD#
BentoDeployment
is a Kubernetes CRD defined by yatai-deployment component.
It is primarily used to describe bento deployments.
Field |
Type |
Description |
---|---|---|
|
|
The version of the schema. Current version is |
|
|
The type of the resource. |
|
|
The metadata of the resource. Refer to the Kubernetes API documentation for the fields of the |
|
|
The name of Bento CR. If this Bento CR not found. yatai-deployment will look for the BentoRequest CR by this name and wait for the BentoRequest CR to generate the Bento CR. required |
|
|
The ingress configuration. |
|
|
Whether the ingress is enabled. |
|
|
The environment variables. |
|
|
The name of the environment variable. |
|
|
The value of the environment variable. |
|
|
The autoscaling configuration for the API server |
|
|
The maximum number of replicas. |
|
|
The minimum number of replicas. |
|
|
The metrics definition |
|
|
The CPU request. |
|
|
The memory request. |
|
|
The list of runners resources configuration. |
|
|
The name of the runner. |
|
|
The maximum number of replicas. |
|
|
The minimum number of replicas. |
|
|
The resources of the runner. |
|
|
The CPU limit. |
|
|
The memory limit. |
|
|
The CPU request. |
|
|
The memory request. |
|
|
The environment variables. |
|
|
The name of the environment variable. |
|
|
The value of the environment variable. |
Example of a BentoDeployment#
apiVersion: serving.yatai.ai/v2alpha1
kind: BentoDeployment
metadata:
name: my-bento-deployment
namespace: my-namespace
spec:
bento: iris-1
ingress:
enabled: true
envs:
- name: foo
value: bar
resources:
limits:
cpu: 2000m
memory: "4Gi"
requests:
cpu: 1000m
memory: "2Gi"
autoscaling:
maxReplicas: 5
minReplicas: 1
metrics:
- type: Resource
resource:
name: cpu
target:
type: Utilization
averageUtilization: 60
runners:
- name: runner1
resources:
limits:
cpu: 2000m
memory: "4Gi"
requests:
cpu: 1000m
memory: "2Gi"
autoscaling:
maxReplicas: 2
minReplicas: 1
- name: runner2
resources:
limits:
cpu: 2000m
memory: "4Gi"
requests:
cpu: 1000m
memory: "2Gi"
autoscaling:
maxReplicas: 4
minReplicas: 1