CloudComputingAndSecurity

Image Object Detection API within a Containerised Environment

A python based web service that accepts images, uses YOLO and OpenCV to process images, and returns a JSON response with a list of detected objects.
Built a Docker Image for the object detection web service.
Created a Kubernetes cluster on a virtual machine (instance) in the Nectar cloud.
Deployed a Kubernetes service to distribute inbound requests among pods that are running the object detection service.
Tested the system under different load and number of pods.

Steps: 1

Create an instance

Create an Ubuntu Instance in any cloud provider: Azure Ubuntu Instance, or EC2-Linux instance. In the video instruction I have used Nectar Cloud Ubuntu Instance

Steps: 2

Install Docker

first of all check if docker is installed or not

From Docker Repository

sudo apt-get update
sudo apt-get install -y apt-transport-https ca-certificates curl software-properties-common
curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo apt-key add -
sudo add-apt-repository "deb [arch=amd64] https://download.docker.com/linux/ubuntu $(lsb_release -cs) stable"
sudo apt-get update
sudo apt-get install -y docker-ce

Starting and Enabling docker service

On Ubuntu 18.04 the service is enabled and running after installation.

sudo systemctl enable docker
sudo systemctl start docker

Running docker commands as non-root user

getent group docker
sudo gpasswd -a <user> docker
getent group docker

Log out and log back in

Step 2 Source : https://github.com/justmeandopensource/learndocker/blob/master/INSTALL_Doc.md

Step: 3

Install GO

wget https://golang.org/dl/go1.15.5.linux-amd64.tar.gz
sudo tar -C /usr/local -xzf go1.15.5.linux-amd64.tar.gz

Set Environment Variable

export PATH=$PATH:/usr/local/go/bin

Step: 4

Install Kind (Kubernetes in docker)

GO111MODULE="on" go get sigs.k8s.io/kind@v0.9.0

Source: https://github.com/kubernetes-sigs/kind After download move go to the /usr/local/go/bin directory

sudo mv go/bin/kind /usr/local/go/bin/

By doing so, you have successfully installed kind. To validate:

kind version

Step: 5

Install Kubectl (Kubernetes command-line tool)

curl -LO https://storage.googleapis.com/kubernetes-release/release/v1.19.0/bin/linux/amd64/kubectl
chmod +x ./kubectl
sudo mv ./kubectl /usr/local/bin/kubectl

to confirm installation

kubectl version --client

Step: 6

Create a Cluster

Clone the repo or download

git clone https://github.com/Farhad-Rezwan/CloudComputingAndSecurity.git

Create cluster

cd CloudComputingAndSecurity
kind create cluster --config kindconfig.yml

To confirm cluster is created

kind get clusters

To confirm configuration of nodes:

kubectl get nodes

Step: 7

Loading Docker image in the cluster:

cd iWebLens/webservice/server/
docker build -t my-custom-image .
kind load docker-image my-custom-image:latest
cd ../../k8sconfig/
kubectl apply -f iweblens_k8s.yml

Step: 10

Allow inbound ports 30001, to make the web-service live in internet

Step: 9

Test on your local machine

Download the client folder from the repo, and inside inputfolder add your image. run the python command

python3 iWebLens_client.py  inputfolder  http://IP_ADDRESS_OF_INSTANCE:30001/api/object_detection 4

Farhad-Rezwan / CloudComputingAndSecurity