Cloud DevOps ND - C4- Microservices at Scale using AWS & Kubernetes - Supporting Material and Project Starter
This repository is associated with Cloud DevOps ND - Course 04 - Microservices at Scale using AWS & Kubernetes. In here, you'll find:
- Supporting material used in the video demonstration in the course
- Starting code for a project, in which you can containerize and deploy a machine learning srevice using Kubernetes.
Download and install the python.
You would require you to install Docker Desktop to create containers for individual microservices. Refer the following links for instructions
- macOS,
- Windows 10 64-bit: Pro, Enterprise, or Education,
- Windows 10 64-bit Home.
- You can find installation instructions for other operating systems at: https://docs.docker.com/install/
You would need to install any one tool for creating a Kubernetes cluster - KubeOne / Minikube / kubectl on top of Docker Desktop:
- Install and Set Up kubectl directly on top of Docker desktop - For Windows/macOS
- Install Minikube - For Linux/macOS
You'll need an AWS account to get started with AWS Lambda, which is a serverless computing platform on cloud.
You may sign up on CircleCI.com with your GitHub credentials.
- Making Change: Create and deploy a serverless lambda function that responds to an input request; this example creates the correct amount of change to make up a value in US dollars.
- Wikipedia Query: Deploy a lambda function that responds to an input, wikipedia page query; this example returns the first sentence of a specific wikipedia page upon being queried.
- Operationalize a Machine Learning Microservice API: Deploy a containerized, machine learning application using Kubernetes.
To run any project code, you'll have to set up a virtual environment with the project dependencies. All of the following instructions are to be completed via a terminal/command line prompt.
These instructions also assume you have git
installed for working with Github from a terminal window, but if you do not, you can download that first from this Github installation page.
Now, you're ready to create your local environment!
- If you haven't already done so, clone the project repository, and navigate to the main project folder.
git clone https://github.com/udacity/DevOps_Microservices.git
cd DevOps_Microservices/project-ml-microservice-kubernetes
- Create (and activate) a new environment, named
.devops
with Python 3. If prompted to proceed with the install(Proceed [y]/n)
type y.
python3 -m venv ~/.devops
source ~/.devops/bin/activate
At this point your command line should look something like: (.devops) <User>:project-ml-microservice-kubernetes<user>$
. The (.devops)
indicates that your environment has been activated, and you can proceed with further package installations.
- Installing dependencies via project
Makefile
. Many of the project dependencies are listed in the filerequirements.txt
; these can be installed usingpip
commands in the providedMakefile
. While in your project directory, type the following command to install these dependencies.
make install
Now most of the .devops
libraries are available to you. There are a couple of other libraries that we'll be using, which can be downloaded as specified, below.
While you still have your .devops
environment activated, you will still need to install:
- Docker
- Hadolint
- Kubernetes (Minikube if you want to run Kubernetes locally)
You will need to use Docker to build and upload a containerized application. If you already have this installed and created a docker account, you may skip this step.
-
You’ll need to create a free docker account, where you’ll choose a unique username and link your email to a docker account. Your username is your unique docker ID.
-
To install the latest version of docker, choose the Community Edition (CE) for your operating system, on docker’s installation site. It is also recommended that you install the latest, stable release:
-
After installation, you can verify that you’ve successfully installed docker by printing its version in your terminal:
docker --version
This project also must pass two lint checks; hadolint
checks the Dockerfile for errors and pylint
checks the app.py
source code for errors.
- Install
hadolint
following the instructions, on hadolint's page:
For Mac:
brew install hadolint
For Windows:
scoop install hadolint
- In your terminal, type:
make lint
to run lint checks on the project code. If you haven’t changed any code, all requirements should be satisfied, and you should see a printed statement that rates your code (and prints out any additional comments):
------------------------------------
Your code has been rated at 10.00/10
That's about it! When working with kubernetes, you may need to install some other libraries, but these instructions will set you up with an environment that can build and deploy Docker containers.