data_infra_repo

As Data infra build part of the "Daas (Data as a service) repo", this project shows how to build DS/DE environments via Docker from scratch. Will focus on : 1) System design by practical using cases 2) Docker, package, and libraries env setting up 3) Test, staging, and product develop/deploy workflow development (CI/CD style maybe)

• Daas (Data as a service) repo : Data infra build -> ETL build -> DS application demo
• Airflow Heroku demo : airflow-heroku-dev
• Mlflow Heroku demo : mlflow-heroku-dev

File Structure

# main projects
├── airflow_in_docker_compose
├── celery_redis_flower_infra
├── deploy_dockerhub.sh
├── hadoop_yarn_spark
├── kafka-zookeeper
├── kafka_zookeeper_redis_infra
├── mysql-master-slave

TODO

• Hadoop
  • hadoop_yarn_spark (batch)
  • hadoop_yarn_spark (stream)
  • hadoop namenode, datanode
  • hadoop_yarn_flink
• Kafka
  • Kafka producer, consumer, zk
  • Kafka mirror
  • Kafka-ELK-DB
• airflow
  • airflow app in docker compose
• DB
  • DB sharding (partition)
  • DB replica
  • DB master-follower
  • DB master-master
  • DB binary stream (kafka) to Bigquery/DW
  • DB binary stream ELK
• Microservice

Test

• Play with Docker

Ref

• EG
  • https://docs.docker.com/compose/gettingstarted/
• ZH
  • https://github.com/twtrubiks/docker-tutorial
  • https://zhuanlan.zhihu.com/p/36071226