Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

A Comparison in Performance Between a Selection of Databases

Ohlsson, Andreas LU and Persson, Mikael LU (2019) In LU-CS-EX 2019-04 EDAM05 20191
Department of Computer Science
Abstract
Choosing the correct database for an application can be a daunting task for an application owner. There is a vast plethora of different options to choose from, where choosing a NoSQL database or a relational database management system built on SQL is the most defining decision. With the ever-increasing amounts of data that is produced, more and more databases are emerging.

In this report we compare Cassandra and MongoDB, two NoSQL databases, with PostgreSQL, a relational database management system, to see which database is most suited to store semi-structured video metadata. The focus of the benchmarks is on the performance of read operations for each database. The tool we are using for comparison is the popular Yahoo! Cloud... (More)
Choosing the correct database for an application can be a daunting task for an application owner. There is a vast plethora of different options to choose from, where choosing a NoSQL database or a relational database management system built on SQL is the most defining decision. With the ever-increasing amounts of data that is produced, more and more databases are emerging.

In this report we compare Cassandra and MongoDB, two NoSQL databases, with PostgreSQL, a relational database management system, to see which database is most suited to store semi-structured video metadata. The focus of the benchmarks is on the performance of read operations for each database. The tool we are using for comparison is the popular Yahoo! Cloud Benchmarking System.

The results when running the workloads from the benchmark tool showed that PostgreSQL performed on average better than MongoDB and Cassandra. Therefore, we recommend Axis to use PostgreSQL for storing their video metadata. (Less)
Popular Abstract (Swedish)
Stora mängder metadata måste sparas för att kunna användas i utvecklingssyfte. Genom att använda sig av databaser, kan metadatan effektivt sparas. Men vilken databas passar bäst för vårt system? Det är vad vår uppsats har svarat på.
Please use this url to cite or link to this publication:
author
Ohlsson, Andreas LU and Persson, Mikael LU
supervisor
organization
course
EDAM05 20191
year
type
H2 - Master's Degree (Two Years)
subject
keywords
Databases, NoSQL, SQL, Benchmarking, Metadata
publication/series
LU-CS-EX 2019-04
report number
LU-CS-EX 2019-04
ISSN
1650-2884
language
English
id
8975288
date added to LUP
2019-09-16 11:02:20
date last changed
2019-09-16 11:02:20
@misc{8975288,
  abstract     = {{Choosing the correct database for an application can be a daunting task for an application owner. There is a vast plethora of different options to choose from, where choosing a NoSQL database or a relational database management system built on SQL is the most defining decision. With the ever-increasing amounts of data that is produced, more and more databases are emerging. 

In this report we compare Cassandra and MongoDB, two NoSQL databases, with PostgreSQL, a relational database management system, to see which database is most suited to store semi-structured video metadata. The focus of the benchmarks is on the performance of read operations for each database. The tool we are using for comparison is the popular Yahoo! Cloud Benchmarking System.

The results when running the workloads from the benchmark tool showed that PostgreSQL performed on average better than MongoDB and Cassandra. Therefore, we recommend Axis to use PostgreSQL for storing their video metadata.}},
  author       = {{Ohlsson, Andreas and Persson, Mikael}},
  issn         = {{1650-2884}},
  language     = {{eng}},
  note         = {{Student Paper}},
  series       = {{LU-CS-EX 2019-04}},
  title        = {{A Comparison in Performance Between a Selection of Databases}},
  year         = {{2019}},
}