Introduction to NoSQL
Learn to query column-oriented, document and key-value data with Snowflake, Postgres JSON and Redis, and to judge which NoSQL database fits a task.
NoSQL databases handle large volumes of unstructured and semi-structured data, and they add ways to store and retrieve that data which a traditional relational database does not offer. You compare column-oriented, document, key-value and graph stores and weigh the trade-offs of each. You then work with Snowflake, Postgres JSON and Redis to see how each one stores records and returns them to the people and processes that use them.
What You Will Achieve
- Judge which NoSQL data store suits a given workload and its downstream data consumers
- Load data into Snowflake tables and tune their design so queries scan less data
- Pull values out of nested JSON documents in Postgres with built-in operators and functions
- Write and read key-value data in Redis from a Python script
Program Overview
NoSQL Basics and Column-oriented Databases
Sets out the four NoSQL families and their pros and cons, then works through building and querying column-oriented tables in Snowflake.
- Tabular and non-tabular NoSQL data stores
- Creating and populating Snowflake tables with COPY INTO
- Micro-partitioning, data clustering and query pruning
- Common table expressions, views and VARIANT data
Document Databases with Postgres JSON
Treats Postgres as a document store, storing semi-structured data and querying it at the top level and inside nested objects.
- JSON and JSONB data in Postgres
- Converting tabular data to JSON
- Top-level and nested JSON queries
- Arrow, hash arrow, #> and #>> operators
Key-value and Graph Databases
Turns to in-memory key-value storage with Redis and Python, and closes with an overview of graph databases.
- Key-value vs. document databases
- Connecting to a Redis cluster
- Storing Python dictionaries with Redis
- Graph database use cases and providers
Certificate Awarded (Paid)
Job Outlook
These skills apply in data engineering, analytics and data science work, wherever teams keep semi-structured records, design warehouse tables for fast queries or rely on an in-memory data store for quick reads.
Good for: Data engineers and analysts who want to see how column-oriented, document and key-value stores behave by querying them in Snowflake, Postgres and Redis.
Keep in mind: Graph databases get only a basic overview in the closing lessons, and document data is handled through Postgres JSON, so the practice is tied to these three tools.
Bottom line: A tool-based look at three NoSQL families, strongest on Snowflake table design and Postgres JSON queries, with graph stores given a brief closing look.
Frequently Asked Questions
Which tools are used for each type of NoSQL database?
Column-oriented work runs in Snowflake, document data is stored and queried with Postgres JSON and JSONB, and key-value data is written and read in Redis. Graph databases are covered as an overview of how they are used and which providers offer them.
How does Snowflake handle semi-structured data in the course?
The Snowflake chapter ends with its semi-structured data types, including VARIANT, and lessons on querying that data, nested values included. Before that, it covers common table expressions and materialized and non-materialized views for building an analytics workflow.
Which Postgres JSON operators does the course teach?
The document chapter works with the arrow and hash arrow operators and with the #> and #>> operators. It also uses built-in functions to extract keys from JSON objects, find the type of JSON values, and extract and manipulate document data.
How is Python used in the course?
In the document chapter, Postgres queries are run from Python with sqlalchemy and pandas. In the key-value chapter, Python connects to a Redis cluster, stores and retrieves key-value data, and saves Python dictionaries in Redis.


There are no reviews yet.