Ingress with Ambassador¶

Seldon Core works well with Ambassador, allowing a single ingress to be used to expose ambassador and running machine learning deployments can then be dynamically exposed through seldon-created ambassador configurations. In this doc we will discuss how your Seldon Deployments are exposed via Ambassador and how you can use both to do various production rollout strategies.

Installing Ambassador¶

You have two options when installing Ambassador:

Option 1: Ambassador API Gateway¶

The Ambassador API Gateway is open source and provides all the functionality of a traditional ingress controller. Follow the Ambassador OSS instructions to install it on your kubernetes cluster.

Using helm the steps can be summarised as

kubectl create namespace ambassador || echo "namespace ambassador exists"

helm repo add datawire https://www.getambassador.io
helm install ambassador datawire/ambassador \
  --set image.repository=quay.io/datawire/ambassador \
  --set enableAES=false \
  --set crds.keep=false \
  --namespace ambassador

Option 2: Ambassador Edge Stack¶

The Ambassador Edge Stack is the easiest way to get started with ambassador. The edgectl install command will provision a load balancer, configure TLS, and provide you with an edgestack.me subdomain. The edgestack.me subdomain allows the Ambassador Edge Stack to automatically provision TLS and HTTPS for a domain name. To install AES using edgectl follow the AES installation instructions.

Once the installation has finished, you can run edgectl login --namespace=ambassador <ambassadorEndpoint> to access the Ambassador Edge Policy Console where you can manage your deployment. The ‘Hosts’ tab will provide information about the domain that Ambassador set up during the installation process.

Ambassador REST¶

Assuming Ambassador is exposed at <ambassadorEndpoint> and with a Seldon deployment name <deploymentName> running in a namespace namespace:

For Seldon Core restricted to a namespace, singleNamespace=true, the endpoints exposed are:

http(s)://<ambassadorEndpoint>/seldon/<deploymentName>/api/v1.0/predictions
http(s)://<ambassadorEndpoint>/seldon/<namespace>/<deploymentName>/api/v1.0/predictions

For Seldon Core running cluster wide, singleNamespace=false, the endpoints exposed are all namespaced:

http(s)://<ambassadorEndpoint>/seldon/<namespace>/<deploymentName>/api/v1.0/predictions

Note here that if you chose to install the Ambassador Edge Stack then you will need to use https and the <AmbassadorEndpoint> referenced above will be the domain name that Ambassador created for you (e.g. random-name-1234.edgestack.me)

Example Curl¶

Ambassador REST¶

If you installed the OSS Ambassador API Gateway, and assuming a Seldon Deployment mymodel with Ambassador exposed on 0.0.0.0:8003 you can send a curl request as follows:

curl -v 0.0.0.0:8003/seldon/mymodel/api/v1.0/predictions -d '{"data":{"names":["a","b"],"tensor":{"shape":[2,2],"values":[0,0,1,1]}}}' -H "Content-Type: application/json"

Alternatively, if you installed the Ambassador Edge Stack, and assuming a Seldon Deployment mymodel with the Ambassador hostname random-hostname-1234.edgestack.me:

curl -v https://random-hostname-1234.edgestack.me/seldon/mymodel/api/v1.0/predictions -d '{"data":{"names":["a","b"],"tensor":{"shape":[2,2],"values":[0,0,1,1]}}}' -H "Content-Type: application/json"

Ambassador Configuration Annotations Reference¶

Annotation	Description
`seldon.io/ambassador-config:<configuration>`	Custom Ambassador Configuration
`seldon.io/ambassador-header:<header>`	The header to add to Ambassador configuration
`seldon.io/ambassador-id:<instance id>`	The instance id to be added to Ambassador `ambassador_id` configuration
`seldon.io/ambassador-regex-header:<regex>`	The regular expression header to use for routing via headers
`seldon.io/ambassador-retries:<number of retries>`	The number of times ambassador will retry request on connect-failure. Default 0. Use custom configuration if more control needed.
`seldon.io/ambassador-service-name:<existing_deployment_name>`	The name of the existing Seldon Deployment for shadow or header based routing
`seldon.io/grpc-timeout: <gRPC read timeout (msecs)>`	gRPC read timeout
`seldon.io/rest-timeout:<REST read timeout (msecs)>`	REST read timeout
`seldon.io/ambassador-circuit-breakers-max-connections:<maximum number of connections>`	The maximum number of connections will make to the Seldon Deployment
`seldon.io/ambassador-circuit-breakers-max-pending-requests:<maximum number of queued requests>`	The maximum number of requests that will be queued while waiting for a connection
`seldon.io/ambassador-circuit-breakers-max-requests:<maximum number of parallel outstanding requests>`	The maximum number of parallel outstanding requests to the Seldon Deployment
`seldon.io/ambassador-circuit-breakers-max-retries:<maximum number of parallel retries>`	The maximum number of parallel retries allowed to the Seldon Deployment

All annotations should be placed in spec.annotations.

See below for details.

Canary Deployments¶

Canary rollouts are available where you wish to push a certain percentage of traffic to a new model to test whether it works ok in production. To add a canary to your SeldonDeployment simply add a new predictor section and set the traffic levels for the main and canary to desired levels. For example:

apiVersion: machinelearning.seldon.io/v1alpha2
kind: SeldonDeployment
metadata:
  name: example
spec:
  name: canary-example
  predictors:
  - componentSpecs:
    - spec:
        containers:
        - image: seldonio/mock_classifier_rest:1.2.1
          name: classifier
    graph:
      children: []
      endpoint:
        type: REST
      name: classifier
      type: MODEL
    name: main
    replicas: 1
    traffic: 75
  - componentSpecs:
    - spec:
        containers:
        - image: seldonio/mock_classifier_rest:1.2.2
          name: classifier
    graph:
      children: []
      endpoint:
        type: REST
      name: classifier
      type: MODEL
    name: canary
    replicas: 1
    traffic: 25

The above example has a “main” predictor with 75% of traffic and a “canary” with 25%.

A worked example for canary deployments is provided.

Shadow Deployments¶

Shadow deployments allow you to send duplicate requests to a parallel deployment but throw away the response. This allows you to test machine learning models under load and compare the results to the live deployment.

Simply set the shadow boolean in your shadow predictor.

A worked example for shadow deployments is provided.

To understand more about the Ambassador configuration for this see their docs on shadow deployments.

Header based Routing¶

Header based routing allows you to route requests to particular Seldon Deployments based on headers in the incoming requests.

You simply need to add some annotations to your Seldon Deployment resource.

seldon.io/ambassador-header:<header> : The header to add to Ambassador configuration
- Example: "seldon.io/ambassador-header":"location: london"
seldon.io/ambassador-regex-header:<header> : The regular expression header to add to Ambassador configuration
- Example: "seldon.io/ambassador-header":"location: lond.*"
seldon.io/ambassador-service-name:<existing_deployment_name> : The name of the existing Seldon Deployment you want to attach to as an alternative mapping for requests.
- Example: "seldon.io/ambassador-service-name":"example"

A worked example for header based routing is provided.

To understand more about the Ambassador configuration for this see their docs on header based routing.

Circuit Breakers¶

By preventing additional connections or requests to an overloaded Seldon Deployment, circuit breakers help improve resilience of your system.

You simply need to add some annotations to your Seldon Deployment resource.

seldon.io/ambassador-circuit-breakers-max-connections:<maximum number of connections> : The maximum number of connections will make to the Seldon Deployment
- Example: "seldon.io/ambassador-circuit-breakers-max-connections":"200"
seldon.io/ambassador-circuit-breakers-max-pending-requests:<maximum number of queued requests> : The maximum number of requests that will be queued while waiting for a connection
- Example: "seldon.io/ambassador-circuit-breakers-max-pending-requests":"100"
seldon.io/ambassador-circuit-breakers-max-requests:<maximum number of parallel outstanding requests> : The maximum number of parallel outstanding requests to the Seldon Deployment
- Example: "seldon.io/ambassador-circuit-breakers-max-requests":"200"
seldon.io/ambassador-circuit-breakers-max-retries:<maximum number of parallel retries> : The maximum number of parallel retries allowed to the Seldon Deployment
- Example: "seldon.io/ambassador-circuit-breakers-max-retries":"3"

A worked example for circuit breakers is provided.

To understand more about the Ambassador configuration for this see their docs on circuit breakers.

Multiple Ambassadors in the same cluster¶

To avoid conflicts in a cluster with multiple ambassadors running, you can add the following annotation to your Seldon Deployment resource.

seldon.io/ambassador-id:<instance id>: The instance id to be added to Ambassador ambassador_id configuration

For example,

apiVersion: machinelearning.seldon.io/v1alpha2
kind: SeldonDeployment
metadata:
  name: multi-ambassadors
spec:
  annotations:
    seldon.io/ambassador-id: my_instance
  name: ambassadors-example

Note that your Ambassador instance must be configured with matching ambassador_id.

See AMBASSADOR_ID for details

Custom Amabassador configuration¶

The above discussed configurations should cover most cases but there maybe a case where you want to have a very particular Ambassador configuration under your control. You can acheieve this by adding your confguration as an annotation to your Seldon Deployment resource.

seldon.io/ambassador-config:<configuration> : The custom ambassador configuration
- Example: "seldon.io/ambassador-config":"apiVersion: ambassador/v1\nkind: Mapping\nname: seldon_example_rest_mapping\nprefix: /mycompany/ml/\nservice: production-model-example.seldon:8000\ntimeout_ms: 3000"

A worked example for custom Ambassador config is provided.