Skip to content

Latest commit

 

History

History
111 lines (101 loc) · 5.81 KB

quick-start-demo.md

File metadata and controls

111 lines (101 loc) · 5.81 KB

Quickest Start

We have created an all-in-one docker image containing everything needed for the demo to run. As long as you have Docker installed, you can immediately start your shifu experience.

  1. Pull and run the docker container:

    docker run --network host -it -v /var/run/docker.sock:/var/run/docker.sock edgehub/demo-image-alpine:v0.0.1

  2. Create kubernetes cluster and start shifu services:

    The script below will create a kubernetes cluster with pre-defined CustomResourceDefinition, as well as starting the minimal required shifu services.

    ./test/scripts/deviceshifu-setup.sh apply

    After the that step, try

    kubectl get pods --all-namespaces

    and we should have the following pods running like this:

    NAMESPACE            NAME                                         READY   STATUS    RESTARTS   AGE
crd-system           crd-controller-manager-7bc78896b9-cpk7d      2/2     Running   0          11m
kube-system          coredns-558bd4d5db-khlqs                     1/1     Running   0          13m
kube-system          coredns-558bd4d5db-w4tvl                     1/1     Running   0          13m
kube-system          etcd-kind-control-plane                      1/1     Running   0          13m
kube-system          kindnet-75547                                1/1     Running   0          13m
kube-system          kube-apiserver-kind-control-plane            1/1     Running   0          13m
kube-system          kube-controller-manager-kind-control-plane   1/1     Running   0          13m
kube-system          kube-proxy-g5kbl                             1/1     Running   0          13m
kube-system          kube-scheduler-kind-control-plane            1/1     Running   0          13m
local-path-storage   local-path-provisioner-547f784dff-wspb2      1/1     Running   0          13m

    we can check the logs to verify the running status:

    kubectl --namespace crd-system logs crd-controller-manager-7bc78896b9-cpk7d -c manager

  3. Start demo deviceShifu (digital twins):

    In directory deviceshifu/examples/demo_device, we have 4 mock devices for shifu to create a deviceShifu (digital twin) for each; all devices have a get_status instruction which returns the current status of the device, such like Busy, Error, Idle, etc. Besides get_status, each mock device has its own instruction:

    • thermometer: a thermometer which reports a temperature, it has an instruction read_value which returns the temperature value read from the thermometer
    • agv: an automated guided vehicle, it has two instructions, it has an instruction get_position which returns the position of the vehicle in the x-y coordinate
    • robotarm: a robot arm used in a lab, it has an insturuction of get_coordinate which returns the position of the robot arm in the x-y-z coordinate
    • plate-reader: a plate with a matrix of blocks where each block contains some liquid, it has an insturuction of get_measurement which returns a matrix of the measurement value of the liquid in each block

    Start the deviceShifu for all 4 devices:

    ./test/scripts/deviceshifu-demo.sh apply edgedevice-thermometer
./test/scripts/deviceshifu-demo.sh apply edgedevice-agv
./test/scripts/deviceshifu-demo.sh apply edgedevice-robot-arm
./test/scripts/deviceshifu-demo.sh apply edgedevice-plate-reader

    We should have the following pods running in devices namespace:

    kubectl get pods --namespace devices

    which should give us an output like this:

    NAME                           READY   STATUS    RESTARTS   AGE
agv-5944698b79-qxdmk           1/1     Running   0          86s
robotarm-5478f86fc8-s5kmg      1/1     Running   0          85s
thermometer-6d6d8f759f-4hd6l   1/1     Running   0          28m
plate-reader-6859f67bc5-htxpp         1/1     Running   0          86s

    We can view the info of a deviceShifu with kubectl describe pods. For example,

    kubectl describe pods edgedevice-thermometer-deployment-b648d5c6c-rf88p

  4. Start a nginx server and communicate with each deviceShifu:

    As the pod IPs are not routable, we can run a simple nginx server, and kubectl exec into the nginx pod. A nginx is provided within the container, so we can just start the pod:

    kubectl run nginx --image=nginx:1.21

    Then we are able to get into the shell of the nginx server:

    kubectl exec -it nginx -- bash

    After that, we can call each deviceShifu's instructions on those mock devices and see the return value. Each deviceShifu has its instructions defined in the configmap file.

    
/ # curl http://edgedevice-thermometer/get_status
Busy
/ # curl http://edgedevice-thermometer/read_value
27
/ # curl http://edgedevice-agv/get_status
Busy
/ # curl http://edgedevice-agv/get_position
xpos: 54, ypos: 34
/ # curl http://edgedevice-robotarm/get_status
Busy
/ # curl http://edgedevice-robotarm/get_coordinate
xpos: 55, ypos: 140, zpos: 135
/ # curl http://edgedevice-plate-reader/get_status
Idle
/ # curl http://edgedevice-plate-reader/get_measurement
0.75 0.50 1.34 0.95 2.79 2.66 2.68 0.59 0.97 0.93 0.70 0.62 
0.61 1.47 1.68 1.65 1.05 1.59 0.78 2.92 1.22 1.12 2.86 0.29 
2.15 2.45 1.99 0.36 1.47 0.18 2.47 0.61 2.43 1.53 0.14 2.41 
2.80 2.49 0.63 2.61 1.09 1.46 0.22 1.99 1.46 2.30 0.51 0.41 
1.24 0.78 0.34 2.83 2.76 1.89 2.64 1.79 1.24 1.68 2.84 2.92 
2.09 2.38 0.02 0.47 0.38 1.62 2.65 0.58 2.17 2.70 0.97 2.18 
1.47 0.66 0.61 0.10 2.91 1.61 0.30 2.21 0.46 1.74 1.62 1.01 
1.28 2.27 1.04 0.44 2.47 1.83 0.59 2.09 1.30 2.24 2.87 2.78