Community

Blog Events Webinars Tutorials Forum
  1. Blog
  2. Events
  3. Webinars
  4. Tutorials
  5. Forum
Create Account
×
Community Blog Serverless Containers and Automatic Scaling Based on Traffic Patterns

Serverless Containers and Automatic Scaling Based on Traffic Patterns

This article explains how we can support Knative-based serverless containers on a managed service mesh technology platform.

By Xining Wang

Serverless and Service Mesh are two popular cloud-native technologies from which customers are exploring how to create more value. As we delve into these solutions with our customers, the problem often arises in the intersection between these two popular technologies and how they complement each other. Can we leverage Service Mesh to protect, observe, and release the Knative serverless applications?

This article attempts to explain how to support Knative-based serverless containers on a managed service mesh technology platform. It also supports traffic pattern-based auto-scaling capabilities, which can be replaced by managed service mesh to simplify the complexity of maintaining the underlying infrastructure for users, so that users can easily build their serverless platforms.

Prerequisites

Before we begin, you should already have Alibaba Cloud Service Mesh and Container Service for Kubernetes services activated on your console.

• Create an Alibaba Cloud Service Mesh (ASM) instance whose Istio version is 1.12.4.50 or later. You can refer to this guide to get started with Service Mesh.

• Create a Container Service for Kubernetes (ACK) cluster. You can refer to this guide to get started with ACK.

• Add an ACK cluster to an ASM instance.

• ASM enables data plane KubeAPI access.

Knative Serving Installation Files

The following table describes the installation files included in Knative Serving:

File Description Dependency Object
serving-crds.yaml Required: Knative Serving Core CRD none
serving-core.yaml Required: Knative Serving core components serving-crds.yaml

In addition, Knative CLI (kn) provides a fast and simple plugin for creating Knative resources, such as Knative services and event sources. You do not need to directly create or modify YAML files. You can refer to this link for details on the installation.

Install the Knative Serving Component

1.  Install the required custom resources by running the following command:

kubectl apply -f  
https://alibabacloudservicemesh.oss-cn-beijing.aliyuncs.com/knative/v0.26/serving-crds.yaml

2.  Install the core components of Knative Serving by running the following command:

kubectl apply -f https://alibabacloudservicemesh.oss-cn-beijing.aliyuncs.com/knative/v0.26/serving-core.yaml

Enable Service Mesh and Integrated Knative

In managed mode, Alibaba Cloud Service Mesh (ASM) and Container Service for Kubernetes managed by ArgoCD are not in the same cluster environment. You need to enable the KubeAPI access capability of ASM to access the Istio resources in the ASM cluster just like the resources in the ACK cluster. In the ASM console, you can enable this capability by referring to this documentation

Install the Knative Istio controller by running the following command:

kubectl apply -f https://alibabacloudservicemesh.oss-cn-beijing.aliyuncs.com/knative/v0.26/net-istio.yaml

Obtain the external IP address by running the following command:

kubectl --namespace istio-system get service istio-ingressgateway

Save it for use in the Configuring DNS section below.

Verify the Installation

Monitor the Knative component until all components show a STATUSofRunning or Completed. You can do this by running the following command and checking the output:

kubectl get pods -n knative-serving

Sample output:

NAME                                    READY   STATUS    RESTARTS   AGE
activator-558bf66d75-svbgz              1/1     Running   0          94m
autoscaler-6fcd9d475-5kgmw              1/1     Running   0          94m
controller-5f98898db-d5zh4              1/1     Running   0          94m
domain-mapping-67d655f47d-wrzz7         1/1     Running   0          94m
domainmapping-webhook-9f59bb774-z792g   1/1     Running   0          94m
net-istio-controller-846c69dbb-qnchg    1/1     Running   0          94m
net-istio-webhook-86cf98b497-vmdqn      1/1     Running   0          94m
webhook-777c5d4548-6tzj6                1/1     Running   0          94m

Deploy the First Knative Service

1.  Deploy a Knative service. This service accepts the environment variable TARGET and prints Hello ${TARGET}!. Copy the following YAML to a file named hello.yaml:

apiVersion: serving.knative.dev/v1
kind: Service
metadata:
  name: helloworld-go
spec:
  template:
    spec:
      containers:
        - image: registry.cn-hangzhou.aliyuncs.com/acs/helloworld-go:160e4dc8
          ports:
            - containerPort: 8080
          env:
            - name: TARGET
              value: "World"

2.  Deploy the Knative service by running the following command:

kubectl apply -f hello.yaml

3.  View the Knative service list by running the following command:

If your curl is used to access the sample application or your Knative application and cannot use the "Magic DNS (sslip.io)" or "Real DNS" methods, there is a temporary method.

After starting the application, get the URL of the application:

kubectl get ksvc

The output should be similar to:

NAME            URL                                        LATESTCREATED         LATESTREADY           READY   REASON
helloworld-go   http://helloworld-go.default.example.com   helloworld-go-00001   helloworld-go-00001   True

Specify the external IP address that the curl connects to the previously mentioned ingress gateway, and specify the host name of the Knative application using the -H "Host:" command line option. For example, if the Network Layer defines your external IP and port as http://39.97.65.87:80, and you want to access helloworld-go of the applications mentioned earlier, use:

curl -H "Host: helloworld-go.default.example.com" http://39.97.65.87:80

For the helloworld-go sample application provided, the default configuration is used, and the output is:

Hello World!

Zoom to Zero

One of the powerful features of Knative Serving is the built-in auto-scaling, also known as auto-scaling. This means that your Knative service only starts your application to perform its work when needed (in this case, say "Hello world!"). Otherwise, it will scale to zero by rotating down and waiting for new requests to enter.

Re-run the above curl command to access the sample application. Now observe the pods to see how they scale to zero after traffic stops flowing to the URL. The output should be similar to:

kubectl get pod -l serving.knative.dev/service=helloworld-go  -w
NAME                                             READY   STATUS    RESTARTS   AGE
helloworld-go-00001-deployment-6f8dfb548-nwfqc   3/3     Running   0          39s
helloworld-go-00001-deployment-6f8dfb548-nwfqc   3/3     Terminating   0          65s
helloworld-go-00001-deployment-6f8dfb548-nwfqc   0/3     Terminating   0          102s
helloworld-go-00001-deployment-6f8dfb548-nwfqc   0/3     Terminating   0          102s
helloworld-go-00001-deployment-6f8dfb548-nwfqc   0/3     Terminating   0          102s

Traffic Splitting

Each time you change the configuration of the Knative service, a new revision is created. When splitting traffic, Knative splits traffic between different versions of your Knative service.

Create a New Revision

Edit an existing hello.yaml file to include the following:

apiVersion: serving.knative.dev/v1
kind: Service
metadata:
  name: helloworld-go
spec:
  template:
    spec:
      containers:
        - image: registry.cn-hangzhou.aliyuncs.com/acs/helloworld-go:160e4dc8
          ports:
            - containerPort: 8080
          env:
            - name: TARGET
              value: "Knative"

Deploy the updated version of the Knative service by running the following command:

kubectl apply -f hello.yaml

Please note that since we are updating the existing Knative Service hello, the URL will not change, but our new revision has the new name hello-00002.

Access the New Version

Specify curl to connect to the external IP address of the previously mentioned ingress gateway and use the-H "Host:" command line option to specify the host name of the Knative application. For example, if the Network Layer defines your external IP and port as http://39.97.65.87:80, and you want to access helloworld-go of the applications mentioned earlier, use:

curl -H "Host: helloworld-go.default.example.com" http://39.97.65.87:80

For the helloworld-go sample application provided, the default configuration is used, and the output is:

Hello World!

List All Revisions

View the revision list by running the following command:

kubectl get revisions

The output should be similar to:

NAME                  CONFIG NAME     K8S SERVICE NAME   GENERATION   READY   REASON   ACTUAL REPLICAS   DESIRED REPLICAS
helloworld-go-00001   helloworld-go                      1            True             0                 0
helloworld-go-00002   helloworld-go                      2            True             0                 0

Split Traffic Between Revisions

Let's split the traffic between the two revisions and add the section to the bottom of the existing file traffic:

apiVersion: serving.knative.dev/v1
kind: Service
metadata:
  name: helloworld-go
spec:
  template:
    spec:
      containers:
        - image: registry.cn-hangzhou.aliyuncs.com/acs/helloworld-go:160e4dc8
          ports:
            - containerPort: 8080
          env:
            - name: TARGET
              value: "Knative"
  traffic:
  - latestRevision: true
    percent: 50
  - latestRevision: false
    percent: 50
    revisionName: hello-00001

Apply YAML by running the following command:

kubectl apply -f hello.yaml

Verify Traffic Splitting

To verify that traffic splitting has been configured correctly, list the revisions again by running the following command:

kn revisions list

The output should be similar to:

NAME                  SERVICE         TRAFFIC   TAGS   GENERATION   AGE     CONDITIONS   READY   REASON
helloworld-go-00002   helloworld-go   50%              2            59s     4 OK / 4     True
helloworld-go-00001   helloworld-go   50%              1            2m58s   4 OK / 4     True

Access the Knative service multiple times in a browser to view the different output provided by each revision. For example, the service URL can be accessed multiple times from a terminal to view the traffic split between revisions.

The output should be similar to:

Hello Knative!
Hello World!
Hello Knative!
Hello World!

Summary

As the industry's first fully managed Istio-compatible service mesh, Alibaba Cloud Service Mesh (ASM) has maintained consistency with the community and industry trends from the beginning of the architecture. The components of the control plane are hosted on the Alibaba Cloud side and are independent of the user clusters on the data side. ASM is customized and implemented based on community Istio. They provide component capabilities for supporting refined traffic management and security management on the managed control surface side. The managed mode decouples the lifecycle management of Istio components from the managed K8s clusters, making the architecture more flexible and improving system scalability.

Through the preceding content, it can be seen that a managed service mesh technology platform supports Knative-based serverless containers. It can simplify the complexity of maintaining the underlying infrastructure and allow users to easily build their own Serverless platforms.

Auto Scaling Container Service Serverless Serverless Container Service Mesh Knative
0 0 0
Share on

Read previous post:

Serverless Application Programming Model in Cloud-Native Scenarios

Read next post:

The Definition of the New Service Mesh-Driven Scenario: AI Model Services - Model Mesh

Alibaba Container Service

120 posts | 26 followers

You may also like

Comments

Alibaba Container Service

120 posts | 26 followers

Related Products

  • Serverless Workflow

    Visualization, O&M-free orchestration, and Coordination of Stateful Application Scenarios

    Learn More

  • Auto Scaling

    Auto Scaling automatically adjusts computing resources based on your business cycle

    Learn More

  • Serverless Application Engine

    Serverless Application Engine (SAE) is the world's first application-oriented serverless PaaS, providing a cost-effective and highly efficient one-stop application hosting solution.

    Learn More

  • Function Compute

    Alibaba Cloud Function Compute is a fully-managed event-driven compute service. It allows you to focus on writing and uploading code without the need to manage infrastructure such as servers.

    Learn More

More Posts by Alibaba Container Service

See All