Analysis of SQL and NoSQL Databases for Big Data Implementation

This report provides a comprehensive comparison between SQL and NoSQL databases, focusing on their suitability for big data applications. It begins with an introduction to both database types, highlighting their core differences in data storage and structure. The report then delves into key aspects such as data sources, dataset types, data characterization, organizational readiness, and affordability, assessing how SQL and NoSQL handle these elements. Transaction processing requirements and data privacy/security issues are also thoroughly examined, with a detailed analysis of the strengths and weaknesses of each database in these areas. The analysis covers the ability of each database to handle structured, unstructured, and semi-structured data, along with security considerations such as encryption and user authentication. Finally, the report concludes with a recommendation on when to choose SQL or NoSQL, based on factors such as the need for data security, processing speed, and scalability. The report emphasizes the importance of understanding organizational needs and data characteristics to make an informed decision when embarking on a big data project.

Running head: BIG DATA
Big Data
Name of the Student
Name of the University
Author Note

1
BIG DATA
Table of Contents
1. Introduction......................................................................................................................2
2. SQL Database and NoSQL Database..............................................................................2
2.1. SQL Database...........................................................................................................2
2.2. NoSQL Database......................................................................................................3
3. Data Sources and Dataset Types Involved in Big data....................................................4
4. Data Characterization......................................................................................................4
5. Organizational readiness and Affordability.....................................................................5
6. Transaction Processing Requirements.............................................................................6
7. Data Privacy and Security Issues in SQL and NoSQL...................................................6
8. SQL or NoSQL?.............................................................................................................7
9. Conclusion.......................................................................................................................9
References..........................................................................................................................10

2
BIG DATA
1. Introduction
A critical decision that is generally faced by the manufacturing companies that are
embarking on a big data project is to identify which database is to be chosen between the SQL
and NoSQL database. Therefore, the storage of big data undergoes a critical decision making
process that involves evaluation of both the databases in order to select an appropriate one.
Although SQL has an impressive record of accomplishment in data storage and data access,
NoSQL is making impressive gains as well. SQL is a relational database. It enables a rigid and
structured way of storing data, much similar to a phone book. Therefore, in SQL the data is
stored in an organized way as a well-designed schema minimizes the redundancy in data. This is
a major advantage of SQL. However, SQL is not suitable for storing unstructured data, and here
the NoSQL database comes into action (Sharma and Dave 2012). NoSQL database does not
contain tables and instead, all the data is stored in a single file. The data can be easily found but
not generally categorized as that of relational database. The report discusses the different
characteristics of the SQL and NoSQL database on basic of the data sources, dataset types,
organizational readiness and affordability.
2. SQL Database and NoSQL Database
2.1. SQL Database
SQL or structure query language provides a structured way of storing and retrieving data
from a database. It is a more secure option of storing data as it allows only the authorized people
to view or access data from the database. The big data type can be structures un-structured and
semi structured as well. SQL database generally stores structured Data in form of tables
containing rows and columns.

3
BIG DATA
2.2. NoSQL Database
NoSQL is a completely different approach for the database design that accommodates a
wide variety of data models, which includes key value document, columnar and graph format as
well. Therefore, NoSQL database supports and stores the unstructured data as well unlike the
SQL database. NoSQL was build for the purpose of integrating a large scale data base clustering
in cloud and web applications. However, with the many benefits offered by the SQL database,
there arise some constraints as well. NoSQL is not as secure as SQL and therefore large scale
organizations such as Google and Amazon uses NoSQL database for focusing on the narrow
operational goals, and use relational database in places where there is a need for high grade data
consistency. NoSQL is beneficial in many cases for its ability to process a large amount of data.
It is a non-relational database and allows a rapid analysis of large amount of data and disparate
data types. Therefore, in course of time, the NoSQL databases have become a first alternative of
the relational database with variety of advantages that includes scalability, availability and fault
tolerance capability.
The NoSQL database has therefore provided a number of competitive advantages in the
different manufacturing industries. There are different types of NoSQL database, which includes
the Graph database, key value store database, column store database and document database. The
major reason for many business and manufacturing industries to adopt a NoSQL database over a
relational database is the benefits it offers, which includes high velocity data processing, variety
of data, capability of storing and processing a large volume of data and the capability of
managing the data complexity (Hammes, Medero and Mitchell 2014). Furthermore, since the
NoSQL environment are built in a distributed architecture the chances of downtime or failure is
negligible.

4
BIG DATA
3. Data Sources and Dataset Types Involved in Big data
The data sources are wide that includes public web, media, social media, data storage,
sensor data, archives and so on. The public web data consists of government, weather, traffic,
economic, census and similar data that are collected in a huge amount every day. Data related to
media consist of images, videos, audios and other media files. Another huge source of data
includes the data collected from different social media on a daily basis. The data storage consists
of information stored on the different databases (Roijackers and Fletcher 2012). Collection of
these data is important to understand the trends and target audience of the manufacturing
industries. However, the collection of data for It can be collected from everywhere and helps
different organizations and industries in many different ways. The huge size of the data makes it
impossible for the big data to be managed by the traditional data tools. Therefore, need and
evaluation of a proper database is essential for proper storage and access of big data.
Both SQL and NoSQL database can be effectively used for data storage and therefore it
becomes difficult for the manufacturing companies to decide which database to choose. Few
companies are better suited for using the SQL database if the primary criterion of the data
security is its protecting the privacy and the confidentiality of the data (Li and Manoharan 2013).
However, since the NoSQL is a hybrid type of database it provides many advantages of SQL
database and therefore, it is increasingly adopted by the different business and industries.
Therefore, it is certainly important to understand the requirements of the company in order to
decide which database would be beneficial to adapt.
4. Data Characterization

5
BIG DATA
Data characterization refers to the summarization of the general features or objects in a
target class. It is an important aspect of the data storage. The different characteristics that defines
the quality of data includes the accuracy and precision of the data, the legitimacy and data
validity, reliability and consistency of the data.
The SQL database provides completeness and comprehensives in storage and access of
data. SQL enables data interaction. It is a declarative query language that provides a structured
solution of the database. In this database, the data are organized in different tables and therefore
data searching and data access follows a proper
The NoSQL database on the other hand is a more informal and unstructured database.
The NoSQL database is a mechanism for data storage and data retrieval that is not modeled in a
tabular relations that is the base of SQL database. NoSQL provides scalability of the data and
therefore it is more preferable for the many manufacturing organization in choosing NoSQL
database over SQL database.
5. Organizational readiness and Affordability
With the advent of big data and data mining many organizations are making use of this
technology to understand the business needs and target markets. Researches proves that data
present in the world doubles in every two year and therefore, proper methodologies for data
storage is essential. The NoSQL database involves different multitude database, and has different
kind of data storage models (Nayak, Poriya and Poojary 2013). SQL is however, more costly to
implement in relation to the NoSQL database. Therefore, the manufacturing organizations whose
main aim of making use of big data is to understand the target audience and industrial trends can
make use of the NoSQL database as it provides a more cost effective solution to the company.

6
BIG DATA
Since Big Data plays an important role in determining the business requirements of the
major leading companies, organizational readiness is not a concern for the adapting to SQL or
NoSQL database. Therefore, it is the requirements of the company and the affordability that is
needed to be considered while making a choice. The NoSQL provides a schema-free design and
breakthrough performance. There is however certain constraints that are associated with the
NoSQL database, which includes the complexity with which the data is stored in the database.
Therefore, if affordability is a criterion for selection of an appropriate database, NoSQL database
will be considered for storing and accessing the data.
6. Transaction Processing Requirements
Transaction processing is used to maintain the integrity of the database by ensuring the
all the SQL operations are either completed entirely or not completed at all. The transaction
processing requirements of the manufacturing companies include the ease of storing and
updating the data. This is offered by both SQL and NoSQL database. However, in SQL
database, the data is needed to be stored in a structured manner, which is a major problem in the
case of NoSQL database as it stores data in an unstructured manner (Pokorny 2013). However,
the data processing speed of is higher in NoSQL database than SQL database.
7. Data Privacy and Security Issues in SQL and NoSQL
There are certain security issues associated with the SQL and NoSQL database for data
storage. However, SQL is more secure than NoSQL. This is because the SQL database provides
encryption of all the data files and only the authenticated users are allowed to access the data.
On the other hand, the user authentication in NoSQL is not enabled by default and has very weak
password storage. It is furthermore, increasingly vulnerable to SQL injections. Data is encrypted

7
BIG DATA
in SQL that helps in maintaining the confidentiality of the data. The NoSQL database lacks the
encryption support for the data files, which is a major drawback of NoSQL database. The data at
rest remains unencrypted in NoSQL database. These security risks are associated with Big data
as it offers the opportunity of access a huge amount of data. Another area of concern in NoSQL
database is the dataflow protection and data governance. Since the data stored in the SQL
database are encrypted, it does not possess the risk of data theft or modification of data while it is
being transferred. However, it is a persistent problem with NoSQL system, as there is no proper
data governance model unlike SQL (Shar and Tan 2013). The use of NoSQL database is
increasing consequently and therefore it definitely needs a proper data governance model.
Therefore, on evaluating both the SQL and NoSQL database on grounds of security and
confidentiality, there is no doubt on the fact that SQL database is more secure than NoSQL
database. If security is a major criterion of evaluation, then SQL database must be the first choice
of the company. However, SQL too have some concerns when evaluated on basis of the use and
access of the big data. Since the manufacturing industries are embarking on big data in order to
enhance their business processes and operation (McCreary and Kelly 2014). Having a tight
security in the data access could therefore prove to be a hindrance for making use of the big data.
The concept of data mining and big data includes the practice of examining a larger and pre
existing database for generating new information. Accessing the information from a SQL
database can be tiresome and difficult and therefore, it would be beneficial for the manufacturing
industries to opt for the NoSQL database for efficient data storage and access.
8. SQL or NoSQL?
It is suggested for the manufacturing industries to consider the NoSQL database because-

8
BIG DATA
1. SQL database is schema oriented, which indicates that the structure of the data should
be known in advance. This might be a problem for storing an unstructured data like big data.
2. SQL databases are prone to serious performance bottlenecks, which can be challenge
for processing unstructured data (Birgen, Preisig and Morud 2014).
3. The data processing speed is high in NoSQL and the cost of data storage is low as
well.
Therefore, it can be said that NoSQL database is preferred over SQL database as it solves
two major problems, scalability and simplified data storage.
9. Conclusion
Therefore, from the above discussion, it can be concluded that the it will be beneficial for
the manufacturing industries to use NoSQL database for their big data projects. NoSQL will help
in storage of the large amount of unstructured data like big data and in a very low cost as
compared to SQL database. The SQL database is although more secure than NoSQL database, it
can act as a hindrance for processing and accessing a large amount of data. Furthermore,
NoSQL database provides a flexible schema structure for better storage and access of huge
amount of data such as big data. NoSQL offers efficient architecture that offers horizontal
scaling. The open source nature of the NoSQL further makes it more cost effective than SQL
database.

9
BIG DATA
References
Birgen, C., Preisig, H. and Morud, J., 2014. SQL vs. NoSQL.
Hammes, D., Medero, H. and Mitchell, H., 2014. Comparison of NoSQL and SQL Databases in
the Cloud. Proceedings of the Southern Association for Information Systems (SAIS), Macon, GA,
pp.21-22.
Li, Y. and Manoharan, S., 2013, August. A performance comparison of SQL and NoSQL
databases. In Communications, computers and signal processing (PACRIM), 2013 IEEE pacific
rim conference on (pp. 15-19). IEEE.
McCreary, D. and Kelly, A., 2014. Making sense of NoSQL. Shelter Island: Manning, pp.19-20.
Nayak, A., Poriya, A. and Poojary, D., 2013. Type of NOSQL databases and its comparison with
relational databases. International Journal of Applied Information Systems, 5(4), pp.16-19.
Pokorny, J., 2013. NoSQL databases: a step to database scalability in web
environment. International Journal of Web Information Systems, 9(1), pp.69-82.
Roijackers, J. and Fletcher, G., 2012. Bridging sql and nosql. Master's thesis, Eindhoven
University of Technology.
Shar, L.K. and Tan, H.B.K., 2013. Defeating SQL injection. Computer, 46(3), pp.69-77.
Sharma, V. and Dave, M., 2012. Sql and nosql databases. International Journal of Advanced
Research in Computer Science and Software Engineering, 2(8).