Course Wiki

Homework — MLOps Zoomcamp Module 6

### Infrastructure-as-Code with Terraform

MLOps Zoomcamp › Module 6: Best Practices › Homework

Notes

More information here.

Part B

Infrastructure-as-Code

with Terraform

!image

Summary

• Setting up a stream-based pipeline infrastructure in AWS, using Terraform
• Project infrastructure modules (AWS): Kinesis Streams (Producer & Consumer), Lambda (Serving API), S3 Bucket (Model artifacts), ECR (Image Registry)

Further info here:

• Concepts of IaC and Terraform
• Setup and Execution

6B.1: Terraform - Introduction

https://www.youtube.com/watch?v=zRcLgT7Qnio&list=PL3MmuxUbc_hIUISrluw_A7wDSmfOhErJK&index=48

• Introduction
• Setup & Pre-Reqs
• Concepts of Terraform and IaC (reference material from previous courses)

6B.2: Terraform - Modules and Outputs variables

https://www.youtube.com/watch?v=-6scXrFcPNk&list=PL3MmuxUbc_hIUISrluw_A7wDSmfOhErJK&index=49

• What are they?
• Creating a Kinesis module

6B.3: Build an e2e workflow for Ride Predictions

https://www.youtube.com/watch?v=JVydd1K6R7M&list=PL3MmuxUbc_hIUISrluw_A7wDSmfOhErJK&index=50

• TF resources for ECR, Lambda, S3

6B.4: Test the pipeline e2e

https://www.youtube.com/watch?v=YWao0rnqVoI&list=PL3MmuxUbc_hIUISrluw_A7wDSmfOhErJK&index=51

• Demo: apply TF to our use-case, manually deploy data dependencies & test
• Recap: IaC, Terraform, next steps

Additional material on understanding Terraform concepts here: Reference Material

CI/CD

with GitHub Actions

!image

Summary

• Automate a complete CI/CD pipeline using GitHub Actions to automatically trigger jobs to build, test, and deploy our service to Lambda for every new commit/code change to our repository.
• The goal of our CI/CD pipeline is to execute tests, build and push container image to a registry, and update our lambda service for every commit to the GitHub repository.

Further info here: Concepts of CI/CD and GitHub Actions

6B.5: CI/CD - Introduction

https://www.youtube.com/watch?v=OMwwZ0Z_cdk&list=PL3MmuxUbc_hIUISrluw_A7wDSmfOhErJK&index=52

• Architecture (Ride Predictions)
• What are GitHub Workflows?

6B.6: Continuous Integration

https://www.youtube.com/watch?v=xkTWF9c33mU&list=PL3MmuxUbc_hIUISrluw_A7wDSmfOhErJK&index=53

• ci-tests.yml
  • Automate sections from tests: Env setup, Unit test, Integration test, Terraform plan
  • Create a CI workflow to trigger on pull-request to develop branch
  • Execute demo

6B.7: Continuous Delivery

https://www.youtube.com/watch?v=jCNxqXCKh2s&list=PL3MmuxUbc_hIUISrluw_A7wDSmfOhErJK&index=54

• cd-deploy.yml
  • Automate sections from tests: Terraform plan, Terraform apply, Docker build & ECR push, Update Lambda config
  • Create a CD workflow to trigger on push to develop branch
  • Execute demo

Alternative CICD Solutions

• Using args and env variables in docker image, and leveraging makefile commands in cicd
  • Check the repo README
  • Using the args Dockerfile
  • Using build args ECR terraform
  • Updating lambda env variables [Post deploy](https://github.com/Nakulbajaj101/mlops-zoomcamp/blob/main/06-best-practices/code-practice/deploy/run_apply_local.sh

Key concepts

• Docker
• Terraform
• CI/CD
• Deployment Automation
• Model Deployment

Related notes

• mlops-m01-2-vm-in-aws
• mlops-m01-homework
• mlops-m02-homework
• mlops-m03-homework
• mlops-m03-using-an-orchestrator
• mlops-m04-mlops-zoomcamp-4-6-batch-scoring-with-mage
• mlops-m04-optional-streaming-deploying-models-with-kinesis
• mlops-m04-three-ways-of-deploying-a-model
• mlops-m04-web-services-deploying-models-with-flask-and-doc
• mlops-m05-debugging-with-test-suites-and-reports
• mlops-m06-code-quality-linting-and-formatting
• mlops-m06-git-pre-commit-hooks
• mlops-m06-integration-tests-with-docker-compose
• mlops-m06-makefiles-and-make
• mlops-m06-testing-cloud-services-with-localstack
• mlops-m06-testing-python-code-with-pytest
• mlops-m07-course-project

Sources

• Video
• Lesson file