Back to search

University of North Florida. School of Computing

Empirical Evaluation of Cloud IAAS Platforms using System-level Benchmarks

Abstract

dc:description.abstract

Cloud Computing is an emerging paradigm in the field of computing where scalable IT enabled capabilities are delivered ‘as-a-service’ using Internet technology. The Cloud industry adopted three basic types of computing service models based on software level abstraction: Infrastructure-as-a-Service (IaaS), Platform-as-a-Service (PaaS), and Software-as-a-Service (SaaS). Infrastructure-as-a-Service allows customers to outsource fundamental computing resources such as servers, networking, storage, as well as services where the provider owns and manages the entire infrastructure. This allows customers to only pay for the resources they consume. In a fast-growing IaaS market with multiple cloud platforms offering IaaS services, the user's decision on the selection of the best IaaS platform is quite challenging. Therefore, it is very important for organizations to evaluate and compare the performance of different IaaS cloud platforms in order to minimize cost and maximize performance. Using a vendor-neutral approach, this research focused on four of the top IaaS cloud platforms- Amazon EC2, Microsoft Azure, Google Compute Engine, and Rackspace cloud services. This research compared the performance of IaaS cloud platforms using system-level parameters including server, file I/O, and network. System-level benchmarking provides an objective comparison of the IaaS cloud platforms from performance perspective. Unixbench, Dbench, and Iperf are the system-level benchmarks chosen to test the performance of the server, file I/O, and network respectively. In order to capture the performance variability, the benchmark tests were performed at different time periods on weekdays and weekends. Each IaaS platform's performance was also tested using various parameters. The benchmark tests conducted on different virtual machine (VM) configurations should help cloud users select the best IaaS platform for their needs. Also, based on their applications' requirements, cloud users should get a clearer picture of which VM configuration they should choose. In addition to the performance evaluation, the price-per-performance value of all the IaaS cloud platforms was also examined.

Degree

thesis:*
Name thesis:degree_name
Master of Science in Computer and Information Sciences (MS)
Discipline thesis:degree_discipline
Computing
Grantor
University of North Florida. School of Computing
Date dc:date.issued
Summer

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Deval, Niharika
  • Deval, Niharika, 1991-
Advisors dc:contributor.advisor
  • Ahuja, Sanjay P.
  • Eggen, Roger
  • Asaithambi, Asai
Contributors dc:contributor
  • Dr. Sanjay P. Ahuja
  • Dr. Roger Eggen
  • Dr. Asai Asaithambi

Subjects

dc:subject × 5

Rights

dc:rights

Identifiers

dc:identifier.*
Repository record dc:identifier
https://digitalcommons.unf.edu/etd/765

Chain of custody

source
Harvested from
University of North Florida
Base URL
digitalcommons.unf.edu/do/oai/
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Deval, Niharika; Deval, Niharika, 1991-. Empirical Evaluation of Cloud IAAS Platforms using System-level Benchmarks. University of North Florida. School of Computing, https://digitalcommons.unf.edu/etd/765