A/B Testing a backend ML model¶

This tutorial describes how to do A/B testing as part of the release of a backend ML model hosted on KServe using the Iter8 SDK. In this tutorial, communication with the model is via gRPC calls.

Install the Iter8 controller¶

Iter8 can be installed and configured to watch resources either in a single namespace (namespace-scoped) or in the whole cluster (cluster-scoped).

helm install --repo https://iter8-tools.github.io/iter8 --version 1.1 iter8 controller

helm install --repo https://iter8-tools.github.io/iter8 --version 1.1 iter8 controller \
--set clusterScoped=true

For additional install options, see Iter8 Installation.

Deploy the sample application¶

A simple sample two-tier application using the Iter8 SDK is provided. Note that only the frontend component uses the Iter8 SDK. Deploy both the frontend and backend components:

Frontend¶

The frontend component uses the Iter8 SDK method Lookup() before each call to the backend (ML model). The frontend uses the returned version number to route the request to the recommended version of backend.

Deploy the frontend:

kubectl create deployment frontend --image=iter8/abn-sample-kserve-grpc-frontend-go:0.17.3
kubectl expose deployment frontend --name=frontend --port=8090

Backend¶

The backend application component is an ML model. Release it using the Iter8 release chart:

cat <<EOF | helm upgrade --install backend --repo https://iter8-tools.github.io/iter8 release --version 1.1 -f -
environment: kserve
application: 
  metadata:
    labels:
      app.kubernetes.io/version: backend
  modelFormat: sklearn
  runtime: kserve-mlserver
  protocolVersion: v2
  ports:
  - containerPort: 9000
    name: h2c
    protocol: TCP
  versions:
  - metadata:
      labels:
        app.kubernetes.io/version: v0
    storageUri: "gs://seldon-models/sklearn/mms/lr_model"
EOF

Wait for the backend model to be ready:

kubectl wait --for condition=ready isvc/backend-0 --timeout=600s

Generate load¶

In one shell, port-forward requests to the frontend component:

kubectl port-forward service/frontend 8090:8090

In another shell, run a script to generate load from multiple users:

curl -s https://raw.githubusercontent.com/iter8-tools/docs/v0.18.3/samples/abn-sample/generate_load.sh | sh -s --

The load generator and sample frontend application outputs the backend that handled each recommendation. With just one version is deployed, all requests are handled by backend-0. In the output you will see something like:

Recommendation: backend-0

Deploy candidate¶

A candidate version of the model can be deployed simply by adding a second version to the list of versions:

cat <<EOF | helm upgrade --install backend --repo https://iter8-tools.github.io/iter8 release --version 1.1 -f -
environment: kserve
application: 
  metadata:
    labels:
      app.kubernetes.io/version: backend
  modelFormat: sklearn
  runtime: kserve-mlserver
  protocolVersion: v2
  ports:
  - containerPort: 9000
    name: h2c
    protocol: TCP
  versions:
  - metadata:
      labels:
        app.kubernetes.io/version: v0
    storageUri: "gs://seldon-models/sklearn/mms/lr_model"
  - metadata:
      labels:
        app.kubernetes.io/version: v1
    storageUri: "gs://seldon-models/sklearn/mms/lr_model"
EOF

Until the candidate version is ready, calls to Lookup() will return only the version index number 0; that is, the first, or primary, version of the model. Once the candidate version is ready, Lookup() will return both 0 and 1, the indices of both versions, so that requests can be distributed across both versions.

Once both backend versions are responding to requests, the output of the load generator will include recommendations from the candidate version. In this example, you should see something like:

Recommendation: backend-1

Compare versions using Grafana¶

Inspect the metrics using Grafana. If Grafana is deployed to your cluster, port-forward requests as follows:

kubectl port-forward service/grafana 3000:3000

Open Grafana in a browser by going to http://localhost:3000 and login. The default username/password are admin/admin.

Add a JSON API data source default/backend with the following parameters:

• URL: http://iter8.default:8080/abnDashboard
• Query string: namespace=default&application=backend

Create a new dashboard by import. Copy and paste the contents of the abn Grafana dashboard into the text box and load it. Associate it with the JSON API data source above.

The Iter8 dashboard allows you to compare the behavior of the two versions of the backend component against each other and select a winner. Since user requests are being sent by the load generation script, the values in the report may change over time. The Iter8 dashboard will look like the following:

Once you identify a winner, it can be promoted, and the candidate version deleted.

Promote candidate¶

The candidate can be promoted by redefining the primary version and removing the candidate:

cat <<EOF | helm upgrade --install backend --repo https://iter8-tools.github.io/iter8 release --version 1.1 -f -
environment: kserve
application: 
  metadata:
    labels:
      app.kubernetes.io/version: backend
  modelFormat: sklearn
  runtime: kserve-mlserver
  protocolVersion: v2
  ports:
  - containerPort: 9000
    name: h2c
    protocol: TCP
  versions:
  - metadata:
      labels:
        app.kubernetes.io/version: v1
    storageUri: "gs://seldon-models/sklearn/mms/lr_model"
EOF

Calls to Lookup() will now recommend that all traffic be sent to the new primary version backend-0 (currently serving the promoted version of the code).

The output of the load generator will again show just backend_0:

Recommendation: backend-0

Cleanup¶

Delete the backend:

helm delete backend

Delete the frontend:

kubectl delete deploy/frontend svc/frontend

Uninstall Iter8 controller:

helm delete iter8

For additional uninstall options, see Iter8 Uninstall.

If you installed Grafana, you can delete it as follows:

kubectl delete svc/grafana deploy/grafana