NoSQL Database Categories: Differences, Examples, and Real-World Uses

This report provides an overview of NoSQL database technology, focusing on the four fundamental categories: key-value stores, wide column-based stores, document-based stores, and graph-based stores. It explains the distinct characteristics of each category, such as the use of hash tables in key-value stores and the semi-structured data handling capabilities of document-based stores. The report highlights the key differences between these categories, including data organization, scalability, and suitability for different types of applications. It also offers practical examples of where these NoSQL databases are used, such as social media sites, e-commerce platforms, and fraud detection systems, illustrating their real-world applications and advantages over traditional relational databases. The report is well-referenced using APA style.

Running head: DATABASE SYSTEMS
1
Database Systems
Name of the Student
Institution Affiliation

DATABASE SYSTEMS 2
Introduction
The NoSQL database technology provides a mechanism for storing and retrieving data
that has been modeled into a structure that does not symbolize any relational tabular. They are
mostly suitable where a large number of the voluminous amount of data need to be stored and
retrieved. The structure is not in a fixed schema and the scalability is easier, (Abramova,
Bernardino & Furtado, 2014).
Basic Category of NoSQL Database
The NoSQL database is usually classified into four main types of NoSQL databases
namely Key-value stores, wide column-based stores, document-based, and graph based.
A key value store usually uses a hash table where there is a specific key for each item in the data.
The wide column-based store usually depicts a structure where each block contains data of a
specific column. Data are usually stored in cell groups of columns instead of rows and are
commonly queried over a large dataset, (Thanh, Thuy, & Huynh, 2018).
The document-based store utilizes semi-structured data to store and record data. It pairs
each key with a complex data structure often referred to as the document. In graphical-based
database nodes and edges are used to store and retrieve data. The nodes are organized in such a
way that their relationship is represented by the use of edges between the nodes.
Difference between the categories
In key value, each data is stored singly with a specific identifier. Data are not grouped as
in the case document based database and column based store. In graph base, there is no table or
column representation can be transformed from one model to the other freely. The value key

DATABASE SYSTEMS 3
database is mainly applicable in cases where data is mainly in a single cohesive value and it is
very fast where retrieving data. The wide column-based is good in storing data whereby the data
in a column usually differ in value in a row which does ensure no spaces are left. In key value,
the developer has to write the code to parse value contrary to the column based where it is in the
form of columns that are used to parse the value in a separate element, (Bugiotti, Cabibbo,
Atzeni & Torlone, 2014). The document-based database has the capability of handling hierarchy
which can hold complex types of data. The graph-based data type can be built on top of one
another which is not possible in the other database types
Example where used
Key value stores are mainly used in social media sites such as Twitter, Facebook,
Pinterest in the storage of comments, session information, storing cookies, followers, and
unfollowers. Also in Pinterest, they are used in storing list for users.The column-based database
is mainly used in systems that require a lot of heavy write requests, it has been used by Spotify
which uses Cassandra to store user profiles attributes and metadata.
The document-based database is usually incorporated in e-commerce systems such as
Alibaba to store customer details and their respective orders. It also being used gaming platforms
such as SEGA games where it handles approximately eleven million user accounts. In addition, it
is being used in blogging and analytical platforms. Graph-based are mainly used to store network
information such as social connections, they have been used in social media sites such as
Instagram and Twitter. They are also used in fraud detection systems.

DATABASE SYSTEMS 4
References
Abramova, V., Bernardino, J., & Furtado, P. (2014). Experimental evaluation of NoSQL
databases. International Journal of Database Management Systems, 6(3), 1.
Bugiotti, F., Cabibbo, L., Atzeni, P., & Torlone, R. (2014, October). Database design for NoSQL
systems. In International Conference on Conceptual Modeling (pp. 223-231). Springer,
Cham.
Thanh, T. M., Thuy, N. H., & Huynh, N. T. (2018, November). Key-value based data hiding
method for NoSQL database. In 2018 10th International Conference on Knowledge and
Systems Engineering (KSE) (pp. 193-197). IEEE.