In order to make the right decision when choosing a data organization system, it is advisable to conduct a comparative analysis.

Key differences between databases and data warehouse:

Data Warehouse

• stores summary data;
• used for data analysis;
• storage of historical and current data;
• information from various sources providing;
• providing of information on general business operations;

Database

• uses detailed data;
• fixation transactions;
• storage of current data;
• collection of data from one source;
• fixation the main day-to-day operations;

Key differences between data mart and data warehouse

Data Mart

• providing of a thematic data subset that was retrieved from the data warehouse (usually less than 100 GB in size);
• is a repository of valuable data for a specific subgroup;
• fast data analysis;
• getting data from the data warehouse;

Data Warehouse

• significantly larger (terabyte or more);
• contains all cleaned data for business units;
• getting data from databases;

Key differences between data lake and data mart

Data Lake

• contains all raw and unfiltered organization data;
• expedient to use for wider and deeper analysis of raw data;
• a complete solution that acts as a data warehouse, database and data mart;
• availability of a central archive where data marts can be stored in different user areas;

Data Mart

• contains filtered and structured data for a specific department;
• allows to quickly and efficiently analyze relevant information;
• is a one-time solution without ETL process;

Key differences between data lake and data warehouse

Data Warehouse

• storage of cleaned data to create structured data models and reports;
• use of ODS from transactional systems;
• intended for users who need to create reports for analytics;

Data Lake

• storage of all data for the organization;
• use of hardware that makes it possible to economically store large amounts of data (terabytes, petabytes);
• extracting data from all data types, including non-traditional data types (web service logs, social media activity, sensor data, etc.);
• designed for deep analysis that goes beyond data scope that is stored in the repository;

Key differences between databases and data mart

Database

• is a transactional data repository (OLTP);
• fixation of all aspects and activities of one subject in particular;
• contains raw data;
• users do not interact with data in databases;
• is the first step in the ETL process;

Data Mart

• is a warehouse of analytical data (OLAP);
• contains data from several subjects;
• contains processed and verified data that simplifies the process of creating reports;
• direct user interaction with data from data marts;
• is  the last step in the ETL process

Key differences between databases and data lake

Database

• fixation of transactional data that are related to one topic;
• stores traditional data (text, numbers);
• does not perform data cleaning, stores raw data;
• exports its data to another process (operational data storage);
• is the first step in the ETL process;

Data Lake

• recording the activity of many databases and other disparate data sources;
• it is possible to store data of any type (pdf-files, images, sound files, etc.);
• stores raw data, however, a data cleansing procedure is implemented;
• performs all data processing (cleansing and aggregation);
• handles all aspects of the ETL process.