A database is an organized collection of information that can be searched, sorted, and updated.

This
data is often stored electronically in a computer system called a database management system (DBMS).
Databases typically organize data in rows and columns for easy processing and retrieval. Oftentimes,
you’ll need to use a programming language, such as structured query language (SQL), to interact with
your database.
Databases are similar to spreadsheets, but there are several key differences. In general, databases are
much larger than spreadsheets and so can store more data, and they allow for multiple users to access
data at the same time. For these reasons, people who work with data, such as data analysts and data
scientists, often work with databases rather than spreadsheets.

Database Types
There are many different approaches to analyzing the different database types
available. The table below provides a general overview of the various types
currently available:

A.Database Model Types

The three general database types based on the model are:
1. Relational database
2. Non-relational database (NoSQL)
3. Object-oriented database
The difference between the models is the way the information looks inside the
database. Consequently, each model type has a different management system
and data relationships.
1. Relational Database
The relational database model is the most extensively used as well as the oldest
database type. The three critical components of a relational database are:
Tables. An entity type with relations.
Rows. Records or instances of an entity type.
Columns. Value attributes of instances.

1
A relational database provides a set of data rows in response to a query. A query
language, most commonly the Structured Query Language or SQL, helps create
these data views.
Relational Database Features
The main features of a relational database are:
ACID compliant. The database retrains integrity while performing transactions.
Range of data types. Provides the capability to store any data as well as carry out
complex queries.
Collaborative. Multiple users are able to access the database and work on the
same project.
Secure. Access is limited or restricted through user permissions.
Stable. Relational databases are well-understood and documented.
What are Relational Databases Used For?
Relational databases are the most implemented database type. There are many use
cases, some of which include:
Online transaction systems. The database supports many users as well as
frequent queries needed in online transactions.
IoT. Relational databases are lightweight and have the processing power needed
for edge computing.
Data warehouses. The critical component of the data warehouse architecture is
storage. Relational databases are easily integrated and optimized for massive
queries from multiple sources.
Most Popular Relational Databases
There are countless commercial as well as open-source databases.
The top ten most popular relational databases are:
1. Oracle 2. MySQL 3. Microsoft SQL Server 4. PostgreSQL 5. IBM Db2 6. SQLite 7.
Microsoft Access 8. MariaDB 9. Hive 10. Microsoft Azure SQL Database

2. Non-relational Database (NoSQL Database)

2
A non-relational database, or NoSQL ("Not Only SQL"), is a type of database that
models and stores data differently from relational databases. Instead of tables, non-
relational databases model relationships between data in an alternative way.

The 4 NoSQL database types are:

1. Document 2. Key-value 3. Column based 4. Graph

Non-relational Database Features

The main features of non-relational databases are:
Flexible. Handling structured, semi-structured, and unstructured data is a breeze
with non-relational database types.
Scalable and responsive. Massive data storage scales well with on-demand
servers and provides quick query responses.
Zero downtime. High availability for minimal downtime due to near real-time data
replication.
Cloud compatible. The scalability of a cloud computing architecture incorporates
perfectly with non-relational databases.
Multiple data structures. Different information types, as well as multi-model
database formats, are available.

What are Non-relational Databases Used For?

Non-relational databases perform best with variable data structures and massive
amounts of data. Some use cases include:
Real-time systems. A non-relational database combines the operational and
analytical database systems into one. Whether feeding operational data
into Hadoop or serving analytics results from Hadoop, non-relational databases
provide the agile real-time experience.
Personalized experience. Elastic scaling accommodates the massive amounts of
data needed for any customized experience.
Fraud detection. High performance is vital in fraud detection. Non-relational
databases are responsive and reliably meet the low latency requirements of
financial systems.

Most Popular Non-Relational Databases

The ten most popular non-relational databases are:
1. MongoDB 2. Redis 3. Cassandra 4. HBase 5. Neo4j 6. Oracle NoSQL 7.
RavenDB 8. Riak 9. OrientDB 10. CouchDB

3
3. Object Database
An object database similarly represents data to objects in object-oriented
programming.
The critical components of an object-oriented database are:
Objects. The basic building blocks for storing information.
Classes. The schema or blueprint for an object.
Methods. Structured behaviors of a class.
Pointers. Access elements of a database and establish relations between objects.

Object databases combine object-oriented programming concepts with database

capabilities.

Object Database Features

The main features of object databases are:
ACID transactions. All transactions are complete without conflicting changes due
to ACID compliance.
Transparent persistency. Object databases integrate seamlessly with object-
oriented programming languages.
Complex and custom data types. User-defined classes allow for custom as well
as complex data types to exist.
Accessible. Data is easy to save and retrieve.
Easier modeling. Real-world problems and information are more closely related to
objects, which makes complex problems easier to model.

What Are Object Databases Used For?

Object databases perform best with complex data types, where one entity includes
a massive amount of information. Some everyday use cases for this database model
type are:
High-performance applications. Applications where fast data retrieval is vital to
benefit from object databases since data is stored and retrieved as-is.
Scientific purposes. Scientific data, as well as calculations, are complex. Storing
complex information and quick retrieval find use in all kinds of scientific disciplines.
Complex data structures. Due to permanent persistence with objects, database
storage and expansion of complex data is accessible, eliminating the need to
rework the database model.

Most Popular Object Databases

Currently, the top ten most popular object databases are:
1. DB4o 2. ObjectStore 3. Matisse 4. Gemstone/S 5. ObjectDB 6.
ObjectDatabase++ 7. Objectivity/DB 8. Versant 9. Perst 10. Jade

B. Database Types Based on Location

Database types also differ based on the physical location of the storage.

4
The two groups based on location are:
1. Centralized databases
2. Distributed databases
Centralized Database
A centralized database is stored as well as managed in a single location. The
information is available through a network. The end-user has access through the
network to the centralized computer, where the stored information resides.

Centralized Database Features

The main features of a centralized database are:
Data integrity. Keeping data in one location maximizes data integrity and reduces
redundancy. Information accuracy and reliability are enhanced.
Security. A single point of location provides only one access point, leading to
increased data safety.
End-user friendly. Data access, as well as updates, are immediate with a
centralized database. A single database design provides simplicity.
Cost-effective. The labor, power supply, and maintenance are all reduced to a
minimum through a centralized system. The database is easier to maintain from an
administration aspect.
Data preservation. A fault-tolerant setup through disaster recovery solutions.

What Are Centralized Databases Used For?

The benefits of a centralized database are most noticeable with large institutions.
Some use cases include:
Enterprise management. Big organizations use centralized databases to get a
better overview of all the information.
Government data. Centralized databases are prevalent in government
organizations. One access point ensures data security.
Schools and universities. Educational institutions make use of centralized
databases. The maintenance is cost-effective, and the information stays accurate.

Distributed Database
Distributed databases store information across different physical sites. The
database resides on multiple CPUs on a single site or spread out across various
locations. Due to the connections between the distributed databases, the
information appears as a single database to end-users.

Distributed Database Features

5
The most exciting features of a distributed database are:
Location independency. The physical location of the database spreads out
across multiple sites.
Query processing distribution. A complex query splits into multiple sites, which
divides the tasks between different CPUs, reducing bottleneck.
Distributed transactions. Multiple storage locations provide a distributed
recovery method. Commit protocols exist in cases of numerous transactions.
Network linking. The distributed databases interlink through a network where
communication happens between the storages as well as with end-users.
Seamless integration. Although not physically connected, distributed database
parts connect into one logical database.
What Are Distributed Databases Used For?
Distributed databases work best in environments with many sectors where
companies should limit the available information to reduce redundancy. Some
examples include:
Large companies. Most company sectors do not need a complete overview of
data. Distributed databases help reduce the redundancy of data with individual
departments.
Global enterprises. Due to location independence, this database type fits well
with companies with multiple sites.

C. Database Types Based on Design

The design of the storage depends on the business objective.
There are two main approaches to database design based on the function are:
1. Operational (transactional) database
2. Analytical database
Although the databases serve a different purpose, incorporating the two together
creates a data warehouse system.
Operational Database
An operational database manages and controls the fundamental operations within a
business. The database is known as an online transaction processing or OLTP
database. The data collected directly from the source in real-time, providing a view
of daily transactions.

Operational Database Features

Operational databases have the following features:
ACID compliant. Preserving the accuracy and integrity of each transaction is
necessary for data organization.
Fast processing. Operational databases require fast processing due to thousands
of simultaneous requests.
Small storage. Transactional information is only stored temporarily. Therefore,
operational databases serve as a steppingstone before the data is archived.
Regular backups. Collecting and storing data requires constant backups,
making legal compliance an essential factor.

Analytical Database
Analytical databases provide a unified view of all data available within a business. A
complete overview of information within a database is essential for planning,

6
reporting, and making decisions. The database is known as an online analytical
processing (OLAP) database.

Analytical Database Features

The features of an analytical database are:
Distributed workload. The data comes from different operational systems
distributed across nodes.
Multi-dimensional. Enterprise information gains dimensionality through data
aggregation and complex queries across databases.
Query performance. Data denormalization improves query performance for time-
intensive actions.
Horizontal scalability. Analytical databases must scale out as the requirements
for an enterprise to grow.
D. Database Types Based on Hosting
There are multiple hosting options for databases.
The two places where an information system resides are:
1. On-premises databases
2. Cloud databases
The notable difference between the two options is the availability of resources when
the database deployment happens.

On-premises Database
An on-premises database resides in-house. All the software, infrastructure as well as
administration needed for support is local. With large-scale enterprises, the storage
grows to a local data center.

On-premises Database Features

The notable features of on-premises databases are:
Security. Due to the infrastructure being in-house, on-premises databases are the
best solution for storing sensitive information.
Control. The enterprise is in complete control over the available information,
providing a high level of regulation and privacy over the data.
Compliance. Regulatory controls, require knowing the location of sensitive data at
any given moment.

Cloud Database
A cloud database is a hosting solution given by a third-party provider. The pay-as-go
solution provides the database-as-a-service, avoiding the need to set up a data
center physically. The agile approach minimizes the initial investments needed to
acquire data space while quickly expanding as more resources are required.

Cloud Database Features

The best features of a cloud database are:
Scalability. Cloud databases are flexible. Increasing or decreasing resources is
quick due to virtualization.
Management flexibility. The provider manages this database type, which in turn
minimizes the management needed from the client. However, there are also options
for outsourcing maintenance.

7
Cost. With a cloud database, you only pay for what you need. The cost of investing
in technical staff, as well as maintenance, is minimized.

E. Database Types Based on Processing Power

The database processing depends on the business model. Choosing the wrong level
of a database system affects the workflow of an organization and team. Most
database vendors offer multiple solutions to database processing. The two main
ones are:
1. Personal database
2. Commercial database
Businesses leverage the power of both depending on the use case.
Personal Database
Personal databases have single-user access and process on low to medium-powered
machines. Simpler database applications benefit from this database type due to the
low cost and maintenance.
Commercial Database
A commercial database has multiple users with various permissions as well as
numerous applications on high-powered machines. High availability commercial
databases are costly and require constant maintenance as well as support.

NOTE: None of the database types are a replacement for a different kind of
database, and using multiple systems provides versatility over data management.