Cloud service discovery and analysis: a unified framework
Date
2017
Authors
Alfazi, Abdullah
Editors
Advisors
Babar, Ali
Journal Title
Journal ISSN
Volume Title
Type:
Theses
Citation
Statement of Responsibility
Conference Name
Abstract
Over the past few years, cloud computing has been more and more attractive as a new
computing paradigm due to high flexibility for provisioning on-demand computing
resources that are used as services through the Internet. The issues around cloud service
discovery have considered by many researchers in the recent years. However,
in cloud computing, with the highly dynamic, distributed, the lack of standardized
description languages, diverse services offered at different levels and non-transparent
nature of cloud services, this research area has gained a significant attention. Robust
cloud service discovery approaches will assist the promotion and growth of cloud
service customers and providers, but will also provide a meaningful contribution to
the acceptance and development of cloud computing. In this dissertation, we have
proposed an automated cloud service discovery approach of cloud services. We have
also conducted extensive experiments to validate our proposed approach. The results
demonstrate the applicability of our approach and its capability of effectively identifying
and categorizing cloud services on the Internet. Firstly, we develop a novel
approach to build cloud service ontology. Cloud service ontology initially is built
based on the National Institute of Standards and Technology (NIST) cloud computing
standard. Then, we add new concepts to ontology by automatically analyzing real
cloud services based on cloud service ontology Algorithm. We also propose cloud
service categorization that use Term Frequency to weigh cloud service ontology concepts
and calculate cosine similarity to measure the similarity between cloud services.
The cloud service categorization algorithm is able to categorize cloud services to clusters for effective categorization of cloud services. In addition, we use Machine
Learning techniques to identify cloud service in real environment. Our cloud service
identifier is built by utilizing cloud service features extracted from the real cloud service
providers. We determine several features such as similarity function, semantic
ontology, cloud service description and cloud services components, to be used effectively
in identifying cloud service on the Web. Also, we build a unified model to
expose the cloud service’s features to a cloud service search user to ease the process of
searching and comparison among a large amount of cloud services by building cloud
service’s profile. Furthermore, we particularly develop a cloud service discovery Engine
that has capability to crawl the Web automatically and collect cloud services.
The collected datasets include meta-data of nearly 7,500 real-world cloud services
providers and nearly 15,000 services (2.45GB). The experimental results show that
our approach i) is able to effectively build automatic cloud service ontology, ii) is
robust in identifying cloud service in real environment and iii) is more scalable in
providing more details about cloud services.
School/Discipline
School of Computer Science
Dissertation Note
Thesis (Ph.D.) -- University of Adelaide, School of Computer Science, 2018
Provenance
This electronic version is made publicly available by the University of Adelaide in accordance with its open access policy for student theses. Copyright in this thesis remains with the author. This thesis may incorporate third party material which has been used by the author pursuant to Fair Dealing exceptions. If you are the owner of any included third party copyright material you wish to be removed from this electronic version, please complete the take down form located at http://www.adelaide.edu.au/legals