Menu

Operational Databases and Data Warehouses

Lesson 1: Operational Databases and Data Warehouses

What is an operational database?

Operational (OLTP – online transaction processing) databases support day-to-day activity:

  • Placing orders
  • Processing payments
  • Updating accounts
  • Booking appointments

They are optimised for fast, reliable, small transactions: many reads and writes of individual records.

What is a data warehouse?

A data warehouse (OLAP – online analytical processing) combines and organises data for reporting and analysis.

  • Stores historical data
  • Combines data from several systems
  • Is structured for queries across time, product, customer, and other dimensions

Popular platforms: Snowflake, Google BigQuery, Amazon Redshift, Databricks, Microsoft Fabric / Azure Synapse.

Star schema – a common warehouse design:

Table type

Contains

Example

Fact tableMeasurable events (numbers)fact_sales: order_id, date_key, product_key, quantity, revenue
Dimension tablesDescriptive contextdim_date, dim_product, dim_customer, dim_region

Related concepts:

Term

Meaning

Data lakeStores raw data in many formats (often cheap object storage)
LakehouseCombines lake storage with warehouse-style querying and management
Data martA smaller, subject-specific warehouse (e.g., marketing only)

Why separate them?

Aspect

Operational database (OLTP)

Data warehouse (OLAP)

PurposeRun the businessAnalyse the business
QueriesMany small reads/writesFewer, large analytical queries
DataCurrent stateHistorical, integrated
DesignNormalised (many related tables)Often denormalised (star schema)
UsersApplications, customersAnalysts, BI tools
Risk if mixedHeavy analysis slows the app–

Heavy analytical queries on an operational database can slow down the live application, so they are usually separated.

Example

System

Holds

Order systemEach order as it happens
Product systemProduct details and prices
Payment systemRefunds
Marketing platformCampaign and traffic sources
WarehouseAll of the above combined → monthly performance report

Key takeaways

  • Know the purpose and refresh timing of every source.
  • A warehouse is optimised for analysis, but its data can still be delayed or contain quality issues.

Data Analytics Tutorial



Learn via Video Course

Data Analysis and Processing with Python using Pandas - Data Science Course Logo

Data Analysis and Processing with Python using Pandas - Data Science Course

5 hrs



Get in Touch For Details