KFServing provides a Kubernetes Custom Resource Definition for serving machine learning (ML) models on arbitrary frameworks. It aims to solve production model serving use cases by providing performant, high abstraction interfaces for common ML frameworks like Tensorflow, XGBoost, ScikitLearn, PyTorch, and ONNX.
It encapsulates the complexity of autoscaling, networking, health checking, and server configuration to bring cutting edge serving features like GPU Autoscaling, Scale to Zero, and Canary Rollouts to your ML deployments. It enables a simple, pluggable, and complete story for Production ML Serving including prediction, pre-processing, post-processing and explainability. KFServing is being used across various organizations.
KFServing Features and Examples
To learn more about KFServing, how to deploy it as part of Kubeflow, how to use various supported features, and how to participate in the KFServing community, please follow the KFServing docs on the Kubeflow Website. Additionally, we have compiled a list of KFServing presentations and demoes to dive through various details.
Kubernetes 1.17 is the minimally recommended version, Knative Serving and Istio should be available on Kubernetes Cluster.
-
Istio: v1.9.0+
- KFServing currently only depends on
Istio Ingress Gatewayto route requests to inference services externally or internally. If you do not needService Mesh, we recommend turning off Istio sidecar injection.
- KFServing currently only depends on
-
Knative Serving: v0.19.0+
- If you are running
Service Meshmode withAuthorizationplease follow knative doc to setup the authorization policies. - If you are looking to use PodSpec fields such as
nodeSelector,affinityortolerationswhich are now supported in the KFServing v1beta1 API spec, you need to turn on the corresponding feature flags in your Knative configuration.
- If you are running
-
Cert Manager: v1.3.0+
- Cert manager is needed to provision KFServing webhook certs for production grade installation, alternatively you can run our self signed certs generation script.
KFServing can be installed standalone if your kubernetes cluster meets the above prerequisites and KFServing controller is deployed in kfserving-system namespace.
TAG=v0.6.0
Install KFServing CRD and Controller
Due to a performance issue applying deeply nested CRDs, please ensure that your kubectl version
fits into one of the following categories to ensure that you have the fix: >=1.16.14,<1.17.0 or >=1.17.11,<1.18.0 or >=1.18.8.
kubectl apply -f https://github.com/kubeflow/kfserving/releases/download/$TAG/kfserving.yamlTo install standalone KFServing on OpenShift Container Platform, please follow the instructions here.
KFServing is installed by default as part of Kubeflow installation and KFServing controller is deployed in kubeflow namespace.
InferenceService in kubeflow namespace which is labelled as control-plane and is system namespace.
Make sure you have kubectl installed.
- If you do not have an existing kubernetes cluster, you can create a quick kubernetes local cluster with kind.
Note that the minimal requirement for running KFServing is 4 cpus and 8Gi memory, so you need to change the docker resource setting to use 4 cpus and 8Gi memory.
kind create clusteralternatively you can use Minikube
minikube start --cpus 4 --memory 8192- Install Istio lean version, Knative Serving, KFServing all in one.(this takes 30s)
./hack/quick_install.shIf the default ingress gateway setup does not fit your need, you can choose to setup a custom ingress gateway
- Configure Custom Ingress Gateway
- In addition you need to update KFServing configmap to use the custom ingress gateway.
- Configure Custom Domain
- Configure HTTPS Connection
Execute the following command to determine if your kubernetes cluster is running in an environment that supports external load balancers
$ kubectl get svc istio-ingressgateway -n istio-system
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
istio-ingressgateway LoadBalancer 172.21.109.129 130.211.10.121 ... 17hIf the EXTERNAL-IP value is set, your environment has an external load balancer that you can use for the ingress gateway.
export INGRESS_HOST=$(kubectl -n istio-system get service istio-ingressgateway -o jsonpath='{.status.loadBalancer.ingress[0].ip}')
export INGRESS_PORT=$(kubectl -n istio-system get service istio-ingressgateway -o jsonpath='{.spec.ports[?(@.name=="http2")].port}')If the EXTERNAL-IP value is none (or perpetually pending), your environment does not provide an external load balancer for the ingress gateway. In this case, you can access the gateway using the service’s node port.
# GKE
export INGRESS_HOST=worker-node-address
# Minikube
export INGRESS_HOST=$(minikube ip)
# Other environment(On Prem)
export INGRESS_HOST=$(kubectl get po -l istio=ingressgateway -n istio-system -o jsonpath='{.items[0].status.hostIP}')
export INGRESS_PORT=$(kubectl -n istio-system get service istio-ingressgateway -o jsonpath='{.spec.ports[?(@.name=="http2")].nodePort}')Alternatively you can do Port Forward for testing purpose
INGRESS_GATEWAY_SERVICE=$(kubectl get svc --namespace istio-system --selector="app=istio-ingressgateway" --output jsonpath='{.items[0].metadata.name}')
kubectl port-forward --namespace istio-system svc/${INGRESS_GATEWAY_SERVICE} 8080:80
# start another terminal
export INGRESS_HOST=localhost
export INGRESS_PORT=8080Expand to see steps for testing the installation!
kubectl get po -n kfserving-system
NAME READY STATUS RESTARTS AGE
kfserving-controller-manager-0 2/2 Running 2 13mPlease refer to our troubleshooting section for recommendations and tips for issues with installation.
API_VERSION=v1beta1
kubectl create namespace kfserving-test
kubectl apply -f docs/samples/${API_VERSION}/sklearn/v1/sklearn.yaml -n kfserving-testkubectl get inferenceservices sklearn-iris -n kfserving-test
NAME URL READY PREV LATEST PREVROLLEDOUTREVISION LATESTREADYREVISION AGE
sklearn-iris http://sklearn-iris.kfserving-test.example.com True 100 sklearn-iris-predictor-default-47q2g 7d23hIf your DNS contains example.com please consult your admin for configuring DNS or using custom domain.
- Curl with real DNS
If you have configured the DNS, you can directly curl the InferenceService with the URL obtained from the status print.
e.g
curl -v http://sklearn-iris.kfserving-test.${CUSTOM_DOMAIN}/v1/models/sklearn-iris:predict -d @./docs/samples/${API_VERSION}/sklearn/v1/iris-input.json
- Curl with magic DNS
If you don't want to go through the trouble to get a real domain, you can instead use "magic" dns xip.io. The key is to get the external IP for your KFServing cluster.
kubectl get svc istio-ingressgateway --namespace istio-system
Look for the EXTERNAL-IP column's value(in this case 35.237.217.209)
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
istio-ingressgateway LoadBalancer 10.51.253.94 35.237.217.209Next step is to setting up the custom domain:
kubectl edit cm config-domain --namespace knative-servingNow in your editor, change example.com to {{external-ip}}.xip.io (make sure to replace {{external-ip}} with the IP you found earlier).
With the change applied you can now directly curl the URL
curl -v http://sklearn-iris.kfserving-test.35.237.217.209.xip.io/v1/models/sklearn-iris:predict -d @./docs/samples/${API_VERSION}/sklearn/v1/iris-input.json- Curl from ingress gateway with HOST Header
If you do not have DNS, you can still curl with the ingress gateway external IP using the HOST Header.
SERVICE_HOSTNAME=$(kubectl get inferenceservice sklearn-iris -n kfserving-test -o jsonpath='{.status.url}' | cut -d "/" -f 3)
curl -v -H "Host: ${SERVICE_HOSTNAME}" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/sklearn-iris:predict -d @./docs/samples/${API_VERSION}/sklearn/v1/iris-input.json- Curl from local cluster gateway
If you are calling from in cluster you can curl with the internal url with host {{InferenceServiceName}}.{{namespace}}
curl -v http://sklearn-iris.kfserving-test/v1/models/sklearn-iris:predict -d @./docs/samples/${API_VERSION}/sklearn/v1/iris-input.json# use kubectl create instead of apply because the job template is using generateName which doesn't work with kubectl apply
kubectl create -f docs/samples/${API_VERSION}/sklearn/v1/perf.yaml -n kfserving-test
# wait the job to be done and check the log
kubectl logs load-test8b58n-rgfxr -n kfserving-test
Requests [total, rate, throughput] 30000, 500.02, 499.99
Duration [total, attack, wait] 1m0s, 59.998s, 3.336ms
Latencies [min, mean, 50, 90, 95, 99, max] 1.743ms, 2.748ms, 2.494ms, 3.363ms, 4.091ms, 7.749ms, 46.354ms
Bytes In [total, mean] 690000, 23.00
Bytes Out [total, mean] 2460000, 82.00
Success [ratio] 100.00%
Status Codes [code:count] 200:30000
Error Set:- Prometheus based monitoring for KFServing
- Metrics driven automated rollouts using Iter8
- Dashboard for ServiceMesh
-
Install the SDK
pip install kfserving -
Check the KFServing SDK documents from here.
-
Follow the example(s) here to use the KFServing SDK to create, rollout, promote, and delete an InferenceService instance.
KFServing Presentations and Demoes
Debug KFServing InferenceService
KFServing benchmark test comparing Knative and Kubernetes Deployment with HPA