An Investigation of NoSQL Database Performance from a

MYSQL Perspective

Introduction:

 This report investigates the two different databases mysql and non-relational databases
nosql What differentiations can be found? What are their pros and cons

 In first we will cover the benefits of each database(sql,nosql) and also we will investigate the
benefits of nosql over mysql

 In second we will run a performance test on each database nosql and mysql, for this test we
use health care data

 The purpose of this investigation is to look into who could benefit from putting their SQL
database in the attic, and to start using a non-relational database instead.

Related Work

Database Technologies:

MongoDB :is a scalable, high-performance, open source, document-oriented database written in C+ +

• Document-Oriented Storage: - Documents (objects) map nicely to programming language data types -
Embedded documents and arrays reduce need for joins - Dynamically-typed (schemaless) for easy
schema evolution - No joins and no multi-document transactions for high performance and easy
scalability
• Replication & High Availability: MongoDB offers the use of Master-slave replication. This means that
you can run "slave"-copies of the master database, and in this way increase the ability to read the
database efficiently. MongoDB's replication also offers a automatic master failover system, electing a
new master if the current one goes down
. • High Performance: - No joins and embedding makes reads and writes fast - Indexes including indexing
of keys from embedded documents and arrays - Optional streaming writes (no acknowledgements)
• Map/Reduce Views and Indexes
Mysql:
MySQL is a database management system (DBMS). A database is a structured collection of data. It might be
anything from a simple shopping list to a picture gallery, to the vast amounts of information in a corporate network.

Measuring Database performance

 Flexibility From a flexibility perspective, structured data doesn't specifically benefit from the use
of a NoSQL DBMS.
  Performance From a performance perspective, structured data can benefit from the scalability
strenghts of NoSQL DBMS if the amount of data stored grows big.

 From a performance perspective, semi-structured and unstructured data does benefit from
the use of a NoSQL DBMS

Methodology
 The research conducted for this thesis will require both qualitative and quantitative research
methodologies.

 To evaluate the process of transitioning from a relational database to a NoSQL database, a case
study will be undertaken that takes an existing application based on a relational database and
converts the application code to use a NoSQL database instead.

 This case study will also be used to assess some of the performance characteristics of the NoSQL
database in comparison to the relational database.

 Thesis Scope A definition of a NoSQL database will be established and the research will be
confined to those databases classed as NoSQL. For reference, the relational database used in
the case study will be MySQL and the NoSQL database will be restricted to a single vendor.

 Other aspects such as backup & recovery, replication, and the robustness of the query language
will not be considered.

Hypothesis

References
[1]: 2011-04-30, http://docs.amazonwebservices.com/AmazonS3/latest/gsg/
[2]: 2011-09-21, http://www.mongodb.org/
[3]: 2011-05-01 https://dev.mysql.com/
[4]: List of NoSQL databases http://nosql-database.org/
[5]: Google Trends NoSQL, http://www.google.com/trends?q=nosql
[6]: Michael Stonebreaker, 2010, SQL Databases v. NoSQL Databases, Communications of the ACM April
Vol 53 NO 4
[7]: Rick Catell, 2011, Scalable SQL and NoSQL Data Stores, ACM SIGMOD Record Volume 39 Issue 4
[8]: Daniel J. Abadi, Peter A. Boncz, Stavros Harizopoulos, 2009, Column-oriented Database Systems,
Proceedings of the VLDB Endowment Volume 2 Issue 2 August
[9]: Natalia Söderberg, Jan Eriksson, 2010, Utredning av NoSQL-databaser för Sogeti I Gävle, Högskolan i
Gävle Akademin för teknik och miljö
[10]: Orend Kai, 2010, Analysis and Classification of NoSQL Databases and Evaluation of their Ability to
Replace an Object-relational Persistance Layer, Technical University Munich Faculty of Informatics
Conclusion