[Course Certification Task] final exam for Cloud Native Edge Computing Course

[RyanZhaoXB]

Hello everyone. This is the final exam for Cloud Native Edge Computing Course.

After 20 open courses, we are finally going to take the final certification exam. Trainees who have completed the exam can fill in materials and apply for KubeEdge Certificate from the community. All the course videos will be uploaded to KubeEdge Course for review.

Here is the final exam.

Design and implement a usage example of KubeEdge. The usage example should focus on one feature of KubeEdge which could be selected from, but not limited to, the following reference list:

• Collect metrics from edge
• Deploy an application which could access the kube-apiserver through metamanager at edge
• Deploy an HTTP server at edge and access it through through beehive and ServiceBus from cloud
• Create RuleEndpoint(rest, eventbus or servicebus) and Rule(rest->eventbus, eventbus->rest or rest->servicebus) in KubeEdge Cluster and make it work for real
• Use KubeEdge to manage the camera at edge and deploy an object recognition application to handle the video stream
• Use Sedna to deploy an inference application to edge
• Use Edgemesh to make two pod in KubeEdge Cluster communicate across edge node
• Use KubeEdge to manage a raspberry pi as an edge node and manage the led as an edge device
• You could also make your own application demo of KubeEdge, which is encouraged.

Please design the scenario, finish the necessary code of demo and provide the specific application documents. After you finish the work, please propose a Pull Request(PR) with the title started with [Course Certification Task] to the example repository of KubeEdge.

After your PR get at least one label of /lgtm, you could apply for KubeEdge Certificate by filling the url of your PR to Application Form. The deadline for application is 22:00 March 22, 2023.

Thanks for pariticipating in the KubeEdge Cloud Native Edge Computing Course.
Best Wishes!

[EnableAsync]

Hello, Thank you for your public course. I have learned a lot about kubeedge from it. And I have implemented an AI model training in the cloud and distributed the trained model to the Raspberry Pi system through KubeEdge. I also created a demo for this system, which controls the on and off of an LED light through face recognition and reports the data to the cloud through MQTT. The specific architecture is shown in the figure.

And, the deployment architecture is shown in the following figure.

The system front-end mainly includes training task management, device management, file management, and service management, as shown in the following figure.

The backend URL is: https://github.com/EnableAsync/cecl-go
The frontend URL is: https://github.com/EnableAsync/cecl-frontend
And the demo written for Raspberry Pi is: https://github.com/EnableAsync/cecl-example

I am not sure if this can meet the requirements for the final assignment, so I will temporarily write the content here. Thank you for reading 😉.

[xingleigao]

kubeedge/examples#129

Hi, I have enhance led-raspberrypi demo. Add web conrtoler for it. And update readme.md.

[dongjiec]

HI,this is Kubeadge Architecture design:

[0LuckyLove0]

Hi，this is collecting metrics from edge

[RyanZhaoXB]

Hi，this is collecting metrics from edge

hello @0LuckyLove0 , could you please propose your deploy.yaml to https://github.com/kubeedge/examples ?