Databases

Relational databases

PostgreSQL

Similar to MySql. Once installed, run psql to access via command line to the instances. An instance of postgresql can have multiple databases.

Another interesting command is pg_dump, it helps you to generate a backup of your tables into a .sql file.

Connection options:
  -h, --host=HOSTNAME      database server host or socket directory (default: "/var/run/postgresql")
  -p, --port=PORT          database server port (default: "5432")
  -U, --username=USERNAME  database user name (default: "<user>")
  -w, --no-password        never prompt for password
  -W, --password           force password prompt (should happen automatically)

Google cloud SQL

From Google cloud you can create an instance of mysql/postgres for you desire database. You can connect via socket or url to database. Online configuration, require code changes.

In addition, you can have a connection using the Cloud SQL Auth proxy

Non-relational databases

Firestore

NoSQL database from Google cloud. python libraries to use and access documents.

Graphs databases

Graphs are a very popular way of representing relationships and connections within your data.

They’re composed of two main components:

  • Vertices : These represent entities/nodes in your data.
  • Edges : These represent connections between entities/nodes.

Graph databases are particularly useful when you have a lot of relationships, because its faster processing of relationships: each node has direct references to its neighbors, so you don’t need to scan the entire dataset to find relationships.

It also allows you to perform complex queries in a more natural way, thanks to graph query language such as Cypher.

Neo4j

Neo4j is a popular graph database that uses the Cypher query language to interact with the database.

Your data is stored without restricting it to a pre-defined model, allowing a very flexible way of thinking about and using it.

Columnar Databases

Columnar databases store data tables by column rather than by row, where the values of a data column are serialized together.

By storing data in columns rather than rows, the database can more precisely access the data it needs to answer a query rather than scanning and discarding unwanted data in rows. Popular in data warehousing and machine learning.

Some columnar databases are:

Columnar database have a series of advantages:

  • Compression: Columnar databases can compress data more effectively than row-based databases because similar data is stored together.
  • Batch transactions.
  • Complex query over data.

References