inferenceservice.yaml

apiVersion: v1
kind: ConfigMap
metadata:
  name: inferenceservice-config
  namespace: kserve
data:
  _example: |-
    ################################
    #                              #
    #    EXAMPLE CONFIGURATION     #
    #                              #
    ################################
    
    # This block is not actually functional configuration,
    # but serves to illustrate the available configuration
    # options and document them in a way that is accessible
    # to users that `kubectl edit` this config map.
    #
    # These sample configuration options may be copied out of
    # this example block and unindented to be in the data block
    # to actually change the configuration.
    
    # ====================================== EXPLAINERS CONFIGURATION ======================================
    # Example
    explainers: |-
      {
          "alibi": {
              "image" : "kserve/alibi-explainer",
              "defaultImageVersion": "latest"
          },
          "art": {
              "image" : "kserve/art-explainer",
              "defaultImageVersion": "latest"
          }
      }
    # Alibi and Art Explainer runtime configuration
     explainers: |-
       {
           # Alibi explainer runtime configuration
           "alibi": {
               # image contains the default Alibi explainer serving runtime image uri.
               "image" : "kserve/alibi-explainer",
               
               # defautltImageVersion contains the Alibi explainer serving runtime default image version.
               "defaultImageVersion": "latest"
           },
           # Art explainer runtime configuration
           "art": {
               # image contains the default Art explainer serving runtime image uri.
               "image" : "kserve/art-explainer",
       
               # defautltImageVersion contains the Art explainer serving runtime default image version.
               "defaultImageVersion": "latest"
           }
       }
     
     # ====================================== STORAGE INITIALIZER CONFIGURATION ======================================
     # Example
     storageInitializer: |-
       {
           "image" : "kserve/storage-initializer:latest",
           "memoryRequest": "100Mi",
           "memoryLimit": "1Gi",
           "cpuRequest": "100m",
           "cpuLimit": "1",
           "enableDirectPvcVolumeMount": false
       }
     storageInitializer: |-
       {
           # image contains the default storage initializer image uri.
           "image" : "kserve/storage-initializer:latest",
           
           # memoryRequest is the requests.memory to set for the storage initializer init container.
           "memoryRequest": "100Mi",
       
            # memoryLimit is the limits.memory to set for the storage initializer init container.
           "memoryLimit": "1Gi",
           
           # cpuRequest is the requests.cpu to set for the storage initializer init container.
           "cpuRequest": "100m",
           
           # cpuLimit is the limits.cpu to set for the storage initializer init container.
           "cpuLimit": "1",
       
           # enableDirectPvcVolumeMount controls whether users can mount pvc volumes directly.
           # if pvc volume is provided in storageuri then the pvc volume is directly mounted to /mnt/models in the user container.
           # rather than symlink it to a shared volume. For more info see https://github.com/kserve/kserve/issues/2737
           "enableDirectPvcVolumeMount": false
       }
     
     # ====================================== CREDENTIALS ======================================
     # Example
     credentials: |-
       {
          "storageSpecSecretName": "storage-config",
          "storageSecretNameAnnotation": "serving.kserve.io/storageSecretName",
          "gcs": {
              "gcsCredentialFileName": "gcloud-application-credentials.json"
          },
          "s3": {
              "s3AccessKeyIDName": "AWS_ACCESS_KEY_ID",
              "s3SecretAccessKeyName": "AWS_SECRET_ACCESS_KEY",
              "s3Endpoint": "",
              "s3UseHttps": "",
              "s3Region": "",
              "s3VerifySSL": "",
              "s3UseVirtualBucket": "",
              "s3UseAnonymousCredential": "",
              "s3CABundle": ""
          }
       }
     # This is a global configuration used for downloading models from the cloud storage.
     # You can override this configuration by specifying the annotations on service account or static secret.
     # https://kserve.github.io/website/master/modelserving/storage/s3/s3/
     # For a quick reference about AWS ENV variables:
     # AWS Cli: https://docs.aws.amazon.com/cli/latest/userguide/cli-configure-envvars.html
     # Boto: https://boto3.amazonaws.com/v1/documentation/api/latest/guide/configuration.html#using-environment-variables
     #
     # The `s3AccessKeyIDName` and `s3SecretAccessKeyName` fields are only used from this configmap when static credentials (IAM User Access Key Secret)
     # are used as the authentication method for AWS S3.
     # The rest of the fields are used in both authentication methods (IAM Role for Service Account & IAM User Access Key Secret) if a non-empty value is provided.
     credentials: |-
       {
          # storageSpecSecretName contains the secret name which has the credentials for downloading the model.
          # This option is used when specifying the storage spec on isvc yaml.
          "storageSpecSecretName": "storage-config",

          # The annotation can be specified on isvc yaml to allow overriding with the secret name reference from the annotation value.
          # When using storageUri the order of the precedence is: secret name reference annotation > secret name references from service account
          # When using storageSpec the order of the precedence is: secret name reference annotation > storageSpecSecretName in configmap

          # Configuration for google cloud storage
          "gcs": {
              # gcsCredentialFileName specifies the filename of the gcs credential
              "gcsCredentialFileName": "gcloud-application-credentials.json"
          },
          
          # Configuration for aws s3 storage. This add the corresponding environmental variables to the storage initializer init container.
          # For more info on s3 storage see https://kserve.github.io/website/master/modelserving/storage/s3/s3/
          "s3": {
              # s3AccessKeyIDName specifies the s3 access key id name
              "s3AccessKeyIDName": "AWS_ACCESS_KEY_ID",
       
              # s3SecretAccessKeyName specifies the s3 secret access key name
              "s3SecretAccessKeyName": "AWS_SECRET_ACCESS_KEY",
              
              # s3Endpoint specifies the s3 endpoint
              "s3Endpoint": "",
              
              # s3UseHttps controls whether to use secure https or unsecure http to download models.
              # Allowed values are 0 and 1.
              "s3UseHttps": "",
       
              # s3Region specifies the region of the bucket.
              "s3Region": "",
              
              # s3VerifySSL controls whether to verify the tls/ssl certificate.
              "s3VerifySSL": "",
              
              # s3UseVirtualBucket configures whether it is a virtual bucket or not.
              "s3UseVirtualBucket": "",
               
              # s3UseAnonymousCredential configures whether to use anonymous credentials to download the model or not.
              "s3UseAnonymousCredential": "",
              
              # s3CABundle specifies the path to a certificate bundle to use for HTTPS certificate validation.
              "s3CABundle": ""
          }
       }
     
     # ====================================== INGRESS CONFIGURATION ======================================
     # Example
     ingress: |-
       {
           "ingressGateway" : "knative-serving/knative-ingress-gateway",
           "ingressService" : "istio-ingressgateway.istio-system.svc.cluster.local",
           "localGateway" : "knative-serving/knative-local-gateway",
           "localGatewayService" : "knative-local-gateway.istio-system.svc.cluster.local",
           "ingressDomain"  : "example.com",
           "ingressClassName" : "istio",
           "domainTemplate": "{{ .Name }}-{{ .Namespace }}.{{ .IngressDomain }}",
           "urlScheme": "http",
           "disableIstioVirtualHost": false
       }
     ingress: |-
       {
           # ingressGateway specifies the ingress gateway to serve external traffic.
           # The gateway should be specified in format <gateway namespace>/<gateway name>
           # NOTE: This configuration only applicable for serverless deployment with Istio configured as network layer.
           "ingressGateway" : "knative-serving/knative-ingress-gateway",
     
           # ingressService specifies the hostname of the ingress service.
           # NOTE: This configuration only applicable for serverless deployment with Istio configured as network layer.
           "ingressService" : "istio-ingressgateway.istio-system.svc.cluster.local",
     
           # localGateway specifies the gateway which handles the network traffic within the cluster.
           # NOTE: This configuration only applicable for serverless deployment with Istio configured as network layer.
           "localGateway" : "knative-serving/knative-local-gateway",
     
           # localGatewayService specifies the hostname of the local gateway service.
           # NOTE: This configuration only applicable for serverless deployment with Istio configured as network layer.
           "localGatewayService" : "knative-local-gateway.istio-system.svc.cluster.local",
     
           # ingressDomain specifies the domain name which is used for creating the url.
           # If ingressDomain is empty then example.com is used as default domain.
           # NOTE: This configuration only applicable for raw deployment.
           "ingressDomain"  : "example.com",
     
           # ingressClassName specifies the ingress controller to use for ingress traffic.
           # This is optional and if omitted the default ingress in the cluster is used.
           # https://kubernetes.io/docs/concepts/services-networking/ingress/#default-ingress-class
           # NOTE: This configuration only applicable for raw deployment.
           "ingressClassName" : "istio",
     
           # domainTemplate specifies the template for generating domain/url for each inference service by combining variable from:
           # Name of the inference service  ( {{ .Name}} )
           # Namespace of the inference service ( {{ .Namespace }} )
           # Annotation of the inference service ( {{ .Annotations.key }} )
           # Label of the inference service ( {{ .Labels.key }} )
           # IngressDomain ( {{ .IngressDomain }} )
           # If domain template is empty the default template {{ .Name }}-{{ .Namespace }}.{{ .IngressDomain }} is used.
           # NOTE: This configuration only applicable for raw deployment.
           "domainTemplate": "{{ .Name }}-{{ .Namespace }}.{{ .IngressDomain }}",
     
           # urlScheme specifies the url scheme to use for inference service and inference graph.
           # If urlScheme is empty then by default http is used.
           "urlScheme": "http",
     
           # disableIstioVirtualHost controls whether to use istio as network layer.
           # By default istio is used as the network layer. When DisableIstioVirtualHost is true, KServe does not
           # create the top level virtual service thus Istio is no longer required for serverless mode.
           # By setting this field to true, user can use other networking layers supported by knative.
           # For more info https://github.com/kserve/kserve/pull/2380, https://kserve.github.io/website/master/admin/serverless/kourier_networking/.
           # NOTE: This configuration is only applicable to serverless deployment.
           "disableIstioVirtualHost": false
     
           # pathTemplate specifies the template for generating path based url for each inference service.
           # The following variables can be used in the template for generating url.
           # Name of the inference service  ( {{ .Name}} )
           # Namespace of the inference service ( {{ .Namespace }} )
           # For more info https://github.com/kserve/kserve/issues/2257.
           # NOTE: This configuration only applicable to serverless deployment.
           "pathTemplate": "/serving/{{ .Namespace }}/{{ .Name }}"
       }
     
     # ====================================== LOGGER CONFIGURATION ======================================
     # Example
     logger: |-
       {
           "image" : "kserve/agent:latest",
           "memoryRequest": "100Mi",
           "memoryLimit": "1Gi",
           "cpuRequest": "100m",
           "cpuLimit": "1",
           "defaultUrl": "http://default-broker"
       }
     logger: |-
       {
           # image contains the default logger image uri.
           "image" : "kserve/agent:latest",
       
           # memoryRequest is the requests.memory to set for the logger container.
           "memoryRequest": "100Mi",
           
           # memoryLimit is the limits.memory to set for the logger container.
           "memoryLimit": "1Gi",
           
           # cpuRequest is the requests.cpu to set for the logger container.
           "cpuRequest": "100m",
           
           # cpuLimit is the limits.cpu to set for the logger container.
           "cpuLimit": "1",
           
           # defaultUrl specifies the default logger url. If logger is not specified in the resource this url is used.
           "defaultUrl": "http://default-broker"
       }
     
     # ====================================== BATCHER CONFIGURATION ======================================
     # Example
     batcher: |-
       {
           "image" : "kserve/agent:latest",
           "memoryRequest": "1Gi",
           "memoryLimit": "1Gi",
           "cpuRequest": "1",
           "cpuLimit": "1"
       }
     batcher: |-
       {
           # image contains the default batcher image uri.
           "image" : "kserve/agent:latest",
           
           # memoryRequest is the requests.memory to set for the batcher container.
           "memoryRequest": "1Gi",
       
           # memoryLimit is the limits.memory to set for the batcher container.
           "memoryLimit": "1Gi",
           
           # cpuRequest is the requests.cpu to set for the batcher container.
           "cpuRequest": "1",
           
           # cpuLimit is the limits.cpu to set for the batcher container.
           "cpuLimit": "1"
       }
     
     # ====================================== AGENT CONFIGURATION ======================================
     # Example
     agent: |-
       {
           "image" : "kserve/agent:latest",
           "memoryRequest": "100Mi",
           "memoryLimit": "1Gi",
           "cpuRequest": "100m",
           "cpuLimit": "1"
       }
     agent: |-
       {
           # image contains the default agent image uri.
           "image" : "kserve/agent:latest",
       
           # memoryRequest is the requests.memory to set for the agent container.
           "memoryRequest": "100Mi",
       
           # memoryLimit is the limits.memory to set for the agent container.
           "memoryLimit": "1Gi",
           
           # cpuRequest is the requests.cpu to set for the agent container.
           "cpuRequest": "100m",
           
           # cpuLimit is the limits.cpu to set for the agent container.
           "cpuLimit": "1"
       }
     
     # ====================================== ROUTER CONFIGURATION ======================================
     # Example
     router: |-
       {
           "image" : "kserve/router:latest",
           "memoryRequest": "100Mi",
           "memoryLimit": "1Gi",
           "cpuRequest": "100m",
           "cpuLimit": "1"
       }
     # router is the implementation of inference graph.
     router: |-
       {
           # image contains the default router image uri.
           "image" : "kserve/router:latest",
           
           # memoryRequest is the requests.memory to set for the router container.
           "memoryRequest": "100Mi",
           
           # memoryLimit is the limits.memory to set for the router container.
           "memoryLimit": "1Gi",
           
           # cpuRequest is the requests.cpu to set for the router container.
           "cpuRequest": "100m",
           
           # cpuLimit is the limits.cpu to set for the router container.
           "cpuLimit": "1"
       }
     
     # ====================================== DEPLOYMENT CONFIGURATION ======================================
     # Example
     deploy: |-
       {
         "defaultDeploymentMode": "Serverless"
       }
     deploy: |-
       {
         # defaultDeploymentMode specifies the default deployment mode of the kserve. The supported values are
         # Serverless, RawDeployment and ModelMesh. Users can override the deployment mode at service level
         # by adding the annotation serving.kserve.io/deploymentMode.For more info on deployment mode visit
         # Serverless https://kserve.github.io/website/master/admin/serverless/serverless/
         # RawDeployment https://kserve.github.io/website/master/admin/kubernetes_deployment/
         # ModelMesh https://kserve.github.io/website/master/admin/modelmesh/
         "defaultDeploymentMode": "Serverless"
       }
     
     # ====================================== METRICS CONFIGURATION ======================================
     # Example
     metricsAggregator: |-
       {
         "enableMetricAggregation": "false",
         "enablePrometheusScraping" : "false"
       }
     # For more info see https://github.com/kserve/kserve/blob/master/qpext/README.md
     metricsAggregator: |-
       {
         # enableMetricAggregation configures metric aggregation annotation. This adds the annotation serving.kserve.io/enable-metric-aggregation to every
         # service with the specified boolean value. If true enables metric aggregation in queue-proxy by setting env vars in the queue proxy container
         # to configure scraping ports.
         "enableMetricAggregation": "false",
         
         # enablePrometheusScraping configures metric aggregation annotation. This adds the annotation serving.kserve.io/enable-metric-aggregation to every
         # service with the specified boolean value. If true, prometheus annotations are added to the pod. If serving.kserve.io/enable-metric-aggregation is false,
         # the prometheus port is set with the default prometheus scraping port 9090, otherwise the prometheus port annotation is set with the metric aggregation port.
         "enablePrometheusScraping" : "false"
       }

  explainers: |-
    {
        "alibi": {
            "image" : "kserve/alibi-explainer",
            "defaultImageVersion": "latest"
        },
        "art": {
            "image" : "kserve/art-explainer",
            "defaultImageVersion": "latest"
        }
    }

  storageInitializer: |-
    {
        "image" : "kserve/storage-initializer:latest",
        "memoryRequest": "100Mi",
        "memoryLimit": "1Gi",
        "cpuRequest": "100m",
        "cpuLimit": "1",
        "enableDirectPvcVolumeMount": false
    }

  credentials: |-
    {
       "storageSpecSecretName": "storage-config",
       "storageSecretNameAnnotation": "serving.kserve.io/storageSecretName",
       "gcs": {
           "gcsCredentialFileName": "gcloud-application-credentials.json"
       },
       "s3": {
           "s3AccessKeyIDName": "AWS_ACCESS_KEY_ID",
           "s3SecretAccessKeyName": "AWS_SECRET_ACCESS_KEY",
           "s3Endpoint": "",
           "s3UseHttps": "",
           "s3Region": "",
           "s3VerifySSL": "",
           "s3UseVirtualBucket": "",
           "s3UseAnonymousCredential": "",
           "s3CABundle": ""
       }
    }

  ingress: |-
    {
        "ingressGateway" : "knative-serving/knative-ingress-gateway",
        "ingressService" : "istio-ingressgateway.istio-system.svc.cluster.local",
        "localGateway" : "knative-serving/knative-local-gateway",
        "localGatewayService" : "knative-local-gateway.istio-system.svc.cluster.local",
        "ingressDomain"  : "example.com",
        "ingressClassName" : "istio",
        "domainTemplate": "{{ .Name }}-{{ .Namespace }}.{{ .IngressDomain }}",
        "urlScheme": "http",
        "disableIstioVirtualHost": false
    }

  logger: |-
    {
        "image" : "kserve/agent:latest",
        "memoryRequest": "100Mi",
        "memoryLimit": "1Gi",
        "cpuRequest": "100m",
        "cpuLimit": "1",
        "defaultUrl": "http://default-broker"
    }

  batcher: |-
    {
        "image" : "kserve/agent:latest",
        "memoryRequest": "1Gi",
        "memoryLimit": "1Gi",
        "cpuRequest": "1",
        "cpuLimit": "1"
    }

  agent: |-
    {
        "image" : "kserve/agent:latest",
        "memoryRequest": "100Mi",
        "memoryLimit": "1Gi",
        "cpuRequest": "100m",
        "cpuLimit": "1"
    }

  router: |-
    {
        "image" : "kserve/router:latest",
        "memoryRequest": "100Mi",
        "memoryLimit": "1Gi",
        "cpuRequest": "100m",
        "cpuLimit": "1"
    }

  deploy: |-
    {
      "defaultDeploymentMode": "Serverless"
    }

  metricsAggregator: |-
    {
      "enableMetricAggregation": "false",
      "enablePrometheusScraping" : "false"
    }