Cloud Service Quality Model: A Cloud Service Quality Model based
on Customer and Provider Perceptions for Cloud Service Mediation
Claudio Giovanoli
Institutue of Information Systems, University of Applied Sciences and Arts Northwestern Switzerland,
Riggenbachstrasse 16, 4600 Olten, Switzerland
Keywords: Cloud Computing, Service Quality, Cloud Service Quality, Quality Attributes.
Abstract: The field of cloud service selection tries to support customers in selecting cloud services based on QoS
attributes. For considering the right, QoS attributes it is necessary to respect the customers and the providers’
perception. This can be made through a Service Quality Model. Thus, this paper introduces a Cloud Service
Quality Model based on a Systematic Literature Review and user interviews as well as providers perceptions.
1 INTRODUCTION
Over the last ten years, the globalization procedure of
business structures has been formed in part through
outsourcing. Outsourcing is a kind of substitution of
internal departments and tasks to third-party vendors
who are typically specialized in certain businesses.
Contracts are regulating supplies and services and the
period of validity between the outsourcing company
and the third-party vendor (Norwood et al., 2006).
Cloud computing can be seen as a stage of IT
outsourcing. The exclusion of internal IT departments
including data centers and complex application
landscapes can be seen as its main drivers.
Soon, companies will need devices connected to
the internet via broadband network access. Other
required services like infrastructure, platforms, and
applications are placed off-premise by cloud service
providers and used on demand. Clients of such cloud
services have no control or influence on the cloud
service providers' IT infrastructure because they just
use the offered service as agreed in SLAs.
Today, companies and organizations planning to
use cloud services are facing a huge number of
different possible cloud solutions. Because of the
immense number of possibilities, it is hard to orient
oneself and find a suitable solution and offer. Cloud
Brokering companies are offering the provision of
optimal service to its customers. This time-
consuming process stands in contrast to the cloud
paradigms of fast provision and on-demand self-
service of a service. Thus, an automated brokerage
approach could leverage the advantages of cloud
computing and increase companies’ agility.
However, before a company can realize these
advantages, a thorough evaluation of the needs,
possible cloud usage scenarios (what type of service
and deployment models will meet), a suitable partner
(who can understand and implement my needs)
should be made in advance. Such a holistic analysis,
however, requires a high use of resources, which
often cannot be guaranteed, especially in the case of
small and medium-sized enterprises, primarily due to
a lack of know-how. There are already tools for
carrying out internal evaluation and procedural
models for the selection of a suitable partner
(provider). However, full consideration can usually
be provided only with the inclusion of consulting
services, which in turn do not pay off especially for
small and medium-sized companies.
2 RELATED WORK
With the growth of cloud service offerings, it has
become increasingly difficult for cloud service
customers to decide which provider can fulfill their
requirements for quality cloud services (Dastjerdi et
al., 2011; Zheng et al., 2013). For example, each
cloud service provider might offer similar services at
different prices and performance levels with different
sets of features (Wibowo and Deng, 2016). However,
while one provider might be cheaper for storage
services, they may be more expensive for
Giovanoli, C.
Cloud Service Quality Model: A Cloud Service Quality Model based on Customer and Provider Perceptions for Cloud Service Mediation.
DOI: 10.5220/0007587502410248
In Proceedings of the 9th International Conference on Cloud Computing and Services Science (CLOSER 2019), pages 241-248
ISBN: 978-989-758-365-0
Copyright
c
2019 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
241
computation. Given the diversity of cloud service
offerings, it is an essential challenge for organizations
to discover suitable cloud providers who can satisfy
their requirements. There may be trade-offs between
different user requirements fulfilled by different
cloud service providers. As a result, it is not sufficient
to discover multiple cloud services. It is important to
determine the most suitable cloud service through an
evaluation for a specific situation (Garg et al., 2013;
Whaiduzzaman et al., 2014; Wibowo and Deng,
2016). The evaluation of available cloud services
concerning a set of specific criteria is complex
(Bryman and Bell, 2015) due to the presence of the
multi-dimensional nature of the evaluation process
and the presence of vagueness of the decision-making
process (Arpaci et al, 2015).
Kritikos and Plexousakis relate the basic web
service discovery and cope with the topic of
requirements for web services discovery (Kritikos
and Plexousakis, 2009). The core of this process is
matchmaking, which enlists the relevant services in
the registry. Afterward, the selection is based on the
ranking approach. First, various services are filtered
and selected as per the user's preference who selects
the options they want to use. This service can be very
likely using Categorization of Super Matches, exact
matches, partially Matches and service that fails. In
this approach, the quality criteria are defined through
literature, based on OWL Q, and consist of
availability, reliability, safety, and security are
considered as criteria. However, according to Hema
Priya and Chandramathi, (Hema Priya and
Chandramathi, 2014), these criteria cannot be
considered all together, which reduces the
opportunities for restrictions. They use numeric QoS
parameters along with their measurement units and
methods in OWL-Q. The criteria reflect users’ needs
and are not considered, and the approach is only
tested prototypically.
A Delphi study conducted by Lang (Lang et al.,
2016) defines the most critical criteria for cloud
provider selection. Through conducting workshops
and panels with industry experts from the cloud
computing area, the authors provided a list of
important selection criteria. This set of criteria
consists of several attributes: certification, contract,
deployment model, flexibility, functionality,
geolocation of service, integration, legal compliance,
monitoring, support, a test of the solution, and
transparency of activities. Using all these criteria can
provide a comprehensive limitation for a cloud
service selection. Nevertheless, they do not offer
measures for their criteria nor a matchmaking method
to prove the approach. This diminishes the
applicability for users as criteria support, tests of
solution and transparency of activities are not easy to
measure, and thus high expertise in each area is
necessary.
A Description Logic-based method proposed by
Dastjerdi supports the QoS-aware discovery of IaaS
web-services and the automatic deployment of
appliances on selected services through a service
(Dastjerdi et al., 2011). The proposed service
matchmaking process has two parts ontologies and
a matchmaking algorithm. The goal of service
matching and five matching operations are first
specified, such as the concepts of exact matching,
plugin matching, non-matching, etc.
In most cases, the project context provides the
language used in the service description. If the
language of the service description is an ontology, the
matchmaker service is based on ontology
fundamentals. In other cases, the service
matchmakers use different mathematical methods.
However, service matchmakers also differ in other
factors: the target service requester, the supported
service layer, their definition for the service
matchmaking process, the types of requirements, and
according to the quality and model used.
Table I examines 20 different service selection
projects. Seven of the selected projects focus on the
selection of web-services, whereas the other projects
focus on Infrastructure or Software Layers. As
functional requirements are underlying on the
systems input/output, most research work is based on
non-functional aspects. Thus, the matchmaking
methods focuses on the matching of non-functional
requirements, mainly QoS aspects.
In the existing approaches, the service description
and quality models stem mostly from the web
services context. Some QoS properties that are
specific to cloud services are not considered, for
example, scalability, elasticity and different price
models. Moreover, some matching approaches do not
provide concrete examples for the service properties
targeted by their service matcher. Considering a
quality model, the approaches are beside
(Repschlaeger et al., 2012; Wang et al., 2014)
linguistic terms most often SMI and OWL-Q.
Whereas OWL-Q appears mostly for web-service
matching, SMI and CFR are used for the selection of
cloud services. As three roles are involved in service
selection, the cloud service customer (CSC), the
cloud service provider (CSP) and the Selector (S)
getting an in-depth look into the research projects,
Table 2 shows that 20 projects are focusing on the
same roles. Somu (Somu et al., 2017) include the CSP
role, beside the CSC as an essential part for building
CLOSER 2019 - 9th International Conference on Cloud Computing and Services Science
242
Table 1: Service Selection Review I.
trust with the customer based on the SMI criteria. All
other projects consider the matchmaking and the
cloud service customer role but excluding the cloud
service provider or for some instances, consider it
only as the provider of datasets. The sources of the
considered requirements of each work are examined:
literature is the dominant source of deriving the used
criteria. Through its Delphi study, Lang (Lang et al.,
2016) conducted panels and interviews with industry
(provider) experts to define their criteria.
II As the cloud service consumer plays a primary
role in the cloud service selection; only two projects
partially consider the needed criteria from an end-
users point of view. In both cases (Lang et al., 2016;
Siegel and Perdue, 2012b) the end-user focus is
represented through the industry experts. All other
projects are not considering any other end-user
derived quality criteria. It can be summarized that the
(i) existing quality models support the selection of
web-services. They can be used for cloud service
selection too, but they do not reflect different aspects
and characteristics of cloud computing (e.g.,
elasticity). Only SMI, CFR and OWL-Q are partially
in favor of the cloud. (ii) The dominantly used non-
functional requirements are derived from academic
literature or only from interviews. There exists no
synthesis of both approaches. (iii) Service selection
consists of the parts customer (CSC), matchmaker
(MM) and provider (CSP). However, the focus is on
CSC and MM, CSPs are neglected. Thus, this work
aims to come up with a Cloud Quality Model, which
reflects cloud characteristics to make cloud services
with similar functional requirements comparable. The
QM considers input from literature, as well as from
cloud services consumers and cloud service
providers. Furthermore, based on the comparability
of the cloud services, the service selection offers also
the opportunity to include cloud service providers to
benchmark their own services.
3 RESEARCH APPROACH
In the design-science paradigm, knowledge and
understanding of a problem domain and its solution
are achieved in the building and application of the
designed artifact. As this research aims to create a
Mediation Broker for evaluating and finding
appropriate cloud services and thus, creates an
artifact, it follows a design-science research strategy.
Regarding the Design Science Research Cycle
(Hevner and Chatterjee, 2010) the application domain
is Cloud Service Selection. Based on a literature
review the relevance of the research is examined. The
Design Cycle consists of two elements to develop and
evaluate. Thus, firstly the development of the cloud
evaluation criteria, the cloud service measures, as
well as the Mediation Broker prototypes, took place.
Followed by the evaluation through a survey and
expert interview for the evaluation criteria and the
cloud service measures and the benchmarking of the
prototypes. There has been extensive and ongoing
research in the field of cloud computing services. As
cloud computing is considered a service, there are
expectations of users which need to be reflected in
offers being provided by the providers (Alabool and
Mahmood, 2013; Garg et al., 2011). In the current
context, the cloud computing services being offered
are not clearly measurable, and often it does not
match up with the expectations of the users, which
creates an environment where a potential user does
not get the confidence in the offering (Buyya, Garg,
and Calheiros, 2011; Garg et al., 2011; Sun et al.,
2014). It then becomes essential that the providers can
define their offering, which helps in mapping the
expectations of the users with the perceived value of
the service provided. Based on these outlines, the
following main and sub-research questions can be
derived:
Research Work Service Layer
Matchmaking and
Selection Context
Type of requirements
Service Quality
Model
Garg et al.
(2011) IaaS Cloud Servives non-functional SMI
Liu et al. (2004) web-services Semantic web-services non-functional WSDL, OWL-S
Sukumar et al.
(2012)
web-services
Web-services from IBM
UDDI Registries
non-functional WSDL, OWL-S
Kritikos et al.
(2009)
web-services
QoS parameters
including parameters
and methods from OWL-
Q
non-functional OWL-Q, OWL-S
Wibowo et al.
(2016)
SaaS Cloud Services non-functional -
Whaiduzzaman
et al. (2014)
IaaS Cloud services functional -
Kang et al.
(2011a)
web-services,
IaaS
Cloud Services functional -
Buyya et al.
(2009)
IaaS, SaaS Cloud services non-functional SMI
Lang et al.
(2016)
IaaS; PaaS,
SaaS Cloud services non-functional -
Sundareswaran
et al. (2012)
IaaS
Cloud Infrastructure
Servuces
non-functional
-
Dastjerdi et al.
(2011)
web-services
Web-services from IBM
UDDI Registries
non-functional
-
Wang (2009) IaaS, SaaS Service Markeptplace non-functional linguistic terms
Sun et al. (2014) IaaS SaaS non-functional -
Zheng et al.
(2013)
SaaS SaaS functional -
Sathya et al.
(2010)
web-services Web-services non-functional WSMO
Shetty et al.
(2015)
SaaS Cloud Services Ranking non-functional -
Siegel et al.
(2012a)
IaaS, PaaS,
SaaS Cloud Services non-functional SMI
Mobedpour et
al. (2013)
- Cloud Service Ranking non-functional -
Somu et al.
(2017)
IaaS, PaaS,
SaaS Cloud Service Ranking non-functional SMI
Raeppschlaeger
et al. (2012)
IaaS
Cloud Service
Evaluation
functional, non
functional
CFR
Cloud Service Quality Model: A Cloud Service Quality Model based on Customer and Provider Perceptions for Cloud Service Mediation
243
Table 2: Service Selection Review.
RQ 1: What is Service Quality; RQ 1.1: What is
Service Quality regarding cloud services
RQ 2: What are Service Quality Models?; RQ 2.1:
Are these SQMs designed for cloud services?; RQ
2.2: What are attributes from the user’s perspective
needs?; RQ 2.3: What are attributes from the
provider’s perspective needs?
4 SERVICE QUALITY MODEL
Kotler and Armstrong (Kotler and Armstrong, 1999)
define service as, “an act of performance that one can
offer to another that is essentially intangible and does
not result in the ownership of anything. Its production
may or may not be tied to a physical product.” It is
conformance of requirements. “Quality is the totality
of features and characteristics of a product or service
that bear on its ability to satisfy stated or implied
needs” (Kotler and Armstrong, 1999).
The quality of service measures how much of the
service provided meets the customers’ expectations.
To measure the quality of intangible services,
researchers usually use the term perceived service
quality. Perceived service quality is the result of
comparing perceptions about the service delivery
process and the actual outcome of the service
(Grönroos, 1984; Wirtz and Lovelock, 2016).
Wang (Wang, 2014) proposed a service quality
management model and service quality evaluation for
maintenance service for cloud computing, a method
based on the SERVQUAL. Using the same
SERVQUAL model, the authors redefined some
quality characteristics, as they argued that
“SERVQUAL is universally applied in the field of
service and cannot reflect the characteristics of
maintenance service for cloud computing.” Based on
the quality management model, this paper proposed a
quality evaluation model using some research
methods, such as the Delphi method. Furthermore,
the paper introduced the application of quality
evaluation by considering an actual case. The essence
of this paper was to help providers improve their
quality management and show them how to deal with
challenges of maintenance service of cloud
computing. This model helps to “solve the problem
underlying in the evaluation of service quality and
inseminate theories and methods for evaluating
service quality.” This paper is more focused on the
provision of quality from the provider’s side, but no
real direct focus on the user’s aspect.
Domínguez-Mayo (Domínguez-Mayo et al.,
2014) proposed a framework and tool to manage
cloud computing service quality. ISO 9000 includes
eight quality management principles, on which to
base an efficient, effective and adaptable quality
management system.
They are applicable throughout industry,
commerce and service sectors: “Customer focus,
leadership, involving people, process approach,
system approach, continual improvement, factual
decision-making, mutually beneficial supplier
relationships, customer requirements, organizations
requirement.” The paper proposed a framework for
managing Cloud Computing service quality between
clients and providers. QuEF (Quality Evaluation
Framework) was developed to manage Model-Driven
Web Development methodologies quality but later
extended to cover the quality management of other
areas like cloud computing. Over time, it has been
improved with the following phases - Strategy Phase,
design phase transition phase, operational phase,
quality continuous improvement phase. The purpose
of the QuEF is to bring about continuous automatic
Research Work
Considered
Roles CSP/
CSC /S
End user
focused
Matchmaking /
Selection
Approach
Garg et al.
(2012) CSC, S literature no AHP
Liu et al. (2004) CSC, S literature no
Semantic
reasoning
Sukumar et al.
(2012)
CSC, S
literature no
Peano space
filling curve
Kritikos et al.
(2009)
CSC, S
literature no
Mixed integer
programing
Wibowo et al.
(2016)
CSC, S
literature no TOPSIS & Fuzzy
Whaiduzzaman
et al. (2014)
CSC, S
literature no AHP
Kang et al.
(2011a)
CSC, S
literature no
Semantic
reasoning
Buyya et al.
(2009)
CSC, S
literature no AHP
Lang et al.
(2016) CSC, S
Panel,
interviews partially -
Sundareswaran
et al. (2012)
CSC, S
literature no
Greedy-
Opitmization
Dastjerdi et al.
(2011)
CSC, S
literature no
Semantic
reasoning
Wang (2009) CSC, S user perception no Fuzzy Logic
Sun et al. (2014) CSC, S literature no AHP
Zheng et al.
(2013)
CSC, S
literature no
Greedy-
Opitmization
Sathya et al.
(2010)
CSC, S
literature no
Shetty et al.
(2015)
CSC, S
literature no AHP
Siegel et al.
(2012a)
CSC, S
interviews partially -
Mobedpour et
al. (2013)
CSC, S
literature no
Ranking similarity
calculation
Somu et al.
(2017)
CSC / CSP
literature no
Raeppschlaeger
et al. (2012)
CSC, S
no
-
CLOSER 2019 - 9th International Conference on Cloud Computing and Services Science
244
improvement by generating checklists,
documentation, thereby automatically evaluating, and
planning in order to improve quality with minimal
effort and time. They did SLR (Systematic Literature
Review), in order to identify gaps in quality
management of Cloud computing, and as per their
study, there are no found works focused on
frameworks, which ensure quality management in
cloud computing service between clients and
providers. The framework further offers a set of tools
to manage quality effectively and efficiently. Further,
more tools help to manage cloud computing quality
between clients and providers by means of e-SCM
(Capability Maturity Model for service) (Domínguez-
Mayo et al., 2015. The article focuses on service
quality, but characteristics of cloud computing
services are not listed, which is also one of the
important aspects of service quality. Zheng, Martin,
Brohman, and Da Xu (Zheng et al., 2016) created a
quality model named CLOUDQUAL for cloud
services. This model is based on SERVQUAL and e-
service quality model. It demonstrates quality
dimensions and metrics for general cloud services.
CLOUDQUAL contains six quality dimensions,
namely, “usability, availability, reliability,
responsiveness, security, and elasticity.” This paper is
focused on validating service quality, and the scope is
limited towards only six dimensions. Moreover,
CLOUDQUAL does not highlight the main
characteristics of cloud computing services, like pay-
per-use, interoperability, etc. A service is defined by
its characteristics and service quality is based on the
characteristics. In this research paper, the scope of the
characteristics is limited and a holistic view, on the
basis of which service quality can be defined are not
covered in this paper. Zheng (Zheng et al., 2013)
proposed a cloud service quality evaluation system
based on five dimensions: “rapport, responsiveness,
reliability, flexibility, security” and extended SaaS-
Qual. The index system proposed external metrics
with the application of SLA in order to measure users’
requirement of service quality for PaaS and SaaS.
The approaches recommended and proposed by
researchers have mainly been for service selection,
but most of them have not focused on service quality,
required for cloud service users and providers.
Another aspect related to service quality is non-
functional attributes like accountability, reliability,
etc., but none of the literature provides a holistic view
of related non-functional attributes of cloud
computing services. The approaches proposed to
cover only one kind of service, or some of them are
useful for either cloud users, providers or
intermediaries. The service quality researches have
been done in other areas of services such as catering,
airline, etc., using the famous SERVQUAL model,
but the same has not been extensively used in cloud
computing services. There is a need to design
classification based on non-functional attributes of
cloud computing services, a scheme, which can
provide qualitative as well as a quantitative view for
providers as well as users of cloud computing
services and measures to evaluate the same.
5 CLOUD SERVICE QUALITY
MODEL
To define a set of quality criteria, to reflect literature,
customer and providers perception, a systematic
literature review based on 46 academic papers, which
have been published in the time frame between 2009
and 2018 was conducted. Out of these papers seven
projects were published between 2009 and 2014, 39
sources were published from 2015 2018. Alabool
and Mahmood reflect in its meta-study already 40
papers considering the most cited criteria from 2009
to 2012 (Alabool and Mahmood, 2013). Their
findings are also included in this research. In the vein
of the search resources, the ScienceDirect (Elsevier),
SpringerLink, ACM Digital Library, and IEEE
Xplore have been considered as the main digital
libraries for cloud computing (Lang et al. 2016; Sun
et al. 2014) for performing search processes. Google
Scholar search engine has been used to find some of
the archival journals, technical reports, and
conference proceedings.
The keywords that have been used to perform a
search over the digital libraries have been selected
based on evaluation theory activities (Lopez, 2000)
that covers the concepts that represent the cloud
evaluation and selection methods domain such as
Cloud Service Evaluation, Cloud Selection Criteria or
Factors, Attributes or Functional Requirements or
Non-Functional Requirements. Based on these
findings in a second step interviews are conducted to
receive in-depth feedback to the quality attributes
(QAs) and its components elaborated within this
work. A short introduction into the general topic of
Cloud Service Selection, Service Quality Models and
the derived QAs gives the interviewees an overview.
The interviews are held through Skype calls and face-
to-face in German and are semi-structured. Semi-
structured interviews are based on a semi-structured
interview guide, which is a schematic presentation of
questions or topics and needs to be explored by the
interviewer (DiCicco-Bloom and Crabtree, 2006).
Cloud Service Quality Model: A Cloud Service Quality Model based on Customer and Provider Perceptions for Cloud Service Mediation
245
This kind of interview offers the advantages of
providing rich data, different ways of data analysis, to
gain more insights about relational aspects and to the
interviewee’s perceptions about the QAs.
The first interviewee is a Service Manager for a
central infrastructure provider and ensures the flow of
information and money between banks, traders,
merchants, investors and service providers
worldwide. The interviewee has worked for Payment
and Card Services, Finance and Insurance,
Healthcare, and Transport industries for more than
ten years and gained experiences in various specialist
fields such as Project & Quality Management, Agile
Service & Product Development, Business
Intelligence, Requirements & Service Management.
The interviewee is a certified cloud expert. The
interviewee is involved in projects for evaluating
cloud services for its company but also in delivering
cloud services to their customers and thus represents
the provider's point of view.
The second interviewee is an experienced Service
Manager with a demonstrated history of working in
the insurance industry. Strong support professional
skilled in Configuration Management, Incident
Management, Service Delivery, Problem
Management, ITIL, and Business Process
Improvement. Currently, she is working for a Swiss
IT-provider for banking services.
The third interviewee is Lead Software Engineer
at an international financial software development
company. Besides his strong skills in developing
cloud services, he gained in his former roles also a
deep insight into IT-Service Management, especially
in the field of Cloud Sourcing for banking institutes.
He represents the customer point of view.
Forth interviewee is Program Test Manager at an
international IT consulting company. She has a strong
background of quality testing for IT services and thus
experiences in quality and metrics. She represents, in
general, the customer's point of view but gives also
general feedback on QAs aspects. Most of the
attributes are recommended as suitable for Quality
Attributes by the interviewees. An exception is the
usability attributed, which is declared as hard to
measure by two interviewees. Additionally, the
questionnaire shows that the derived attributes are at
least suitable for a Cloud Service Quality Model.
Thus, except the attribute usability, the other
attributes are considered as Quality Attributes.
Besides these QAs, the interviews and survey
show that there is a need for additional attributes.
They see the attributes Compliance and Geo-
Location as important criteria while considering the
service quality. Customizing, Reputation, Costs per
Figure 1: Quality Attributes Validation Results.
Costumer and additional Costs are attributes, which
must be considered from the customers and providers
point of view. Based on these findings the categories
and QAs for a Cloud Service Quality Model are
derived in Fig.1. For defining the Cloud Service
Quality Model as a next step, the identified attributes
now are collocated to the different categories.
Therefore, the reviewed attributes from the SLR are
used, and additionally, the additional attributes form
the interviews and survey. As the additional attributes
consist of 50% attributes which are only named once,
only the attributes with at least 25 % frequency (three
nominations) are considered. In total these attributes
are, excluding the attributes which are now a
category: cost, additional costs, costs per user,
availability, scalability, elasticity, interoperability,
portability, customizing, reputation, compliance,
reliability, assurance, number of active users,
certificates, geo-location. As all attributes assigned to
one category, the category of Security, Performance,
and Support to not inherited any attributes. As
mentioned before these groups can have different
attributes and can be generalized through different
attributes (e.g., Security and Privacy). Within the
interviews that for example for the category of
Support metric can be the number of incidents per
year and the average support time. Derived from these
metrics the attributes are Support duration and
Incidents. For the category of Security, the triad of
confidentiality, integrity, and availability can be
considered. As availability reflects an own category,
the preferred attributes are Data confidentiality and
integrity. The Trusted Cloud Label (Verein
Kompetenznetzwerk Trusted Cloud eV, 2016) and
CLOSER 2019 - 9th International Conference on Cloud Computing and Services Science
246
the Service Measurement Index (Siegel and Perdue,
2012b) are selection models that can be used to
measure cloud services according to given criteria.
Both are defining also the Security Management
System (if in place or not) as an important criterion.
Thus, these three attributes are considered as
attributes for the Security category.
As the metrices and measures are broadly
discussed in the literature already, existing metrics
and measures can be considered. For example, the
measurements of the Trusted Cloud label and SMI are
considered as guidelines (where applicable) because
both tend to develop a comprehensive criteria catalog
which covers the defined evaluation criteria within
this work. Furthermore, both approaches are ensuring
that the criteria are suitable to request and analyzing
them in the context of self-service and self-test from
the provider (Siegel and Perdue, 2012b; Verein
Kompetenznetzwerk Trusted Cloud eV, 2016) which
is in the alignment of this work.
6 VALIDATION
For validating the findings, a panel discussion with 21
cloud experts shows the suitability of the QAs for
cloud service selection.
The validation shows that there is a space for
additional attributes for cloud service quality besides
the traditional literature derived attributes. The
attributes derived from providers and customers view
have in general suitability or acceptance over more
than 60% even if attributes like Additional Costs and
Active users are also seen as not important. This
discrepancy lies in the drawback of this work, which
is the limited number of interviews held with
customers and users. Additionally, more interviews
could have led also to additional attributes which now
are not considered.
7 CONCLUSION
Literature and a survey have shown that the process
of finding a suitable cloud service is not trivial. Small
businesses often do not have the knowledge to define
their requirements and find a suitable cloud service.
As literature describes, there are already many
research initiatives that have been or are still in
progress. However, they usually focus on a specific
domain, such as matching, service selection, service
description, or are applicable only to a service or
deployment model. As the concept of service quality
is still not widely prevalent in the cloud computing
services, this study investigates on the service quality
of cloud services, which can be used for cloud service
selection. Thus, following a design science research
approach, a list of the most common cited cloud
service quality attributes has been identified. Based
on these literature derived attributes, the cloud
customer’s and cloud provider’s perception was
collected. Within interviews and a questionnaire, the
topic has been discussed and further attributes were
identified. In a next step, the attributes supported the
creation of according categories. Furthermore, simple
metrics have been identified, where applicable, to
derive a Cloud Service Quality Model.
REFERENCES
Alabool, H. M. and A. K. Mahmood. 2013. “Review on
Cloud Service Evaluation and Selection Methods.”
2013 International Conference on Research and
Innovation in Information Systems (ICRIIS) 2013:61
66.
Arpaci, Ibrahim, Kerem Kilicer, and Salih Bardakci. 2015.
“Effects of Security and Privacy Concerns on
Educational Use of Cloud Services.” Computers in
Human Behavior 45:9398.
Bryman, Alan and Emma Bell. 2015. Business Research
Methods. 4th ed. New York: Oxford University Press.
Buyya, Rajkumar, Saurabh Kumar Garg, and Rodrigo N.
Calheiros. 2011. “SLA-Oriented Resource
Provisioning for Cloud Computing: Challenges,
Architecture, and Solutions.” Pp. 110 in Proceedings
of the 2011 International Conference on Cloud and
Service Computing, CSC ’11. Washington, DC, USA:
IEEE Computer Society.
Dastjerdi, Amir Vahid, Saurabh Kumar Garg, and
Rajkumar Buyya. 2011. “QoS-Aware Deployment of
Network of Virtual Appliances across Multiple
Clouds.” Pp. 415–23 in Proceedings - 2011 3rd IEEE
International Conference on Cloud Computing
Technology and Science, CloudCom 2011. IEEE.
DiCicco-Bloom, Barbara and Benjamin F. Crabtree. 2006.
“The Qualitative Research Interview.” Medical
Education 40(4):31421.
Domínguez-Mayo, Francisco José et al. 2015. A Framework
and Tool to Manage Cloud Computing Service Quality.”
Software Quality Journal 23(4):595625.
Garg, Saurabh Kumar, Christian Vecchiola, and Rajkumar
Buyya. 2013. “Mandi: A Market Exchange for Trading
Utility and Cloud Computing Services.” Journal of
Supercomputing 64(3):115374.
Garg, Saurabh Kumar, Steve Versteeg, and Rajkumar
Buyya. 2011. “SMICloud: A Framework for
Comparing and Ranking Cloud Services.” Proceedings
- 2011 4th IEEE International Conference on Utility
and Cloud Computing, UCC 2011 (Vm):21018.
Cloud Service Quality Model: A Cloud Service Quality Model based on Customer and Provider Perceptions for Cloud Service Mediation
247
Grönroos, Christian. 1984. “A Service Quality Model and
Its Marketing Implications.” European Journal of
Marketing 18(4):3644.
Hema Priya, N. and S. Chandramathi. 2014. “QoS Based
Optimal Selection of Web Services Using Fuzzy
Logic.” edited by S. Fong. Journal of Emerging
Technologies in Web Intelligence 6(3):33139.
Hevner, Alan and Samir Chatterjee. 2010. Design Research
in Information Systems. Springer US.
Kang, Jaeyong and Kwang Mong Sim. 2011. “Cloudle: An
Ontology-Enhanced Cloud Service Search Engine.” Pp.
41627 in Lecture Notes in Computer Science
(including subseries Lecture Notes in Artificial
Intelligence and Lecture Notes in Bioinformatics), vol.
6724 LNCS.
Kotler, Philip and Gary Armstrong. 1999. “Marketing
Principles.” Parsaian, Ali, Tehran, Adabestan Jahan e
Nou Publication 1389.
Kritikos, Kyriakos et al. 2013. A Survey on Service
Quality Description.” ACM Computing Surveys.
Kritikos, Kyriakos and Dimitris Plexousakis. 2009.
“Requirements for QoS-Based Web Service
Description and Discovery.” Pp. 320–37 in IEEE
Transactions on Services Computing, vol. 2. IEEE.
Lang, Michael, Manuel Wiesche, and Helmut Krcmar.
2016. “What Are the Most Important Criteria for Cloud
Service Provider Selection? A Delphi Study.” Pp. 1–17
in Twenty Fourth European Conference on Information
Systems (ECIS 2016). Istanbul: AIS Electronic Library.
Retrieved (http://aisel.aisnet.org/ecis2016_rp/119).
Li, Zheng, Ranjan,Rajiv, O’Brien, Liam, Zheng Li, Liam
O’Brien, and Rajiv Ranjan. 2016. “Cloud Service
Evaluation.” Pp. 349–60 in Encyclopedia of Cloud
Computing. John Wiley & Sons, Ltd.
Liu, Yutu, Anne H. Ngu, and Liang Z. Zeng. 2004. “QoS
Computation and Policing in Dynamic Web Service
Selection.” in Proceedings of the 13th international
World Wide Web conference on Alternate track papers
& posters - WWW Alt. ’04.
Lopez, Marta. 2000. An Evaluation Theory Perspective of
the Architecture Tradeoff Analysis Method (ATAM).
Pittsburgh.
Mingrui Sun, Tianyi Zang, Xiaofei Xu, and Rongjie Wang.
2013. “Consumer-Centered Cloud Services Selection
Using AHP.” Pp. 1–6 in 2013 International Conference
on Service Sciences (ICSS). Shenzhen: IEEE. Retrieved
(http://ieeexplore.ieee.org/document/6519752/).
Mobedpour, Delnavaz and Chen Ding. 2013. “User-
Centered Design of a QoS-Based Web Service
Selection System.” Service Oriented Computing and
Applications 7(2):11727.
Norwood, Janet et al. 2006. Off-Shoring: An Elusive
Phenomenon. National Academy of Public
Administration.
Repschlaeger, Jonas, Ruediger Zarnekow, Stefan Wind,
and Turowski Klaus. 2012. “Cloud Requirement
Framework: Requirements and Evaluation Criteria to
Adopt Cloud Solutions.” 20th European Conference on
Information Systems.
Sathya, M. and M. Swarnamugi. 2010. “Evaluation of QoS
Based Web- Service Selection Techniques for Service
Composition.” International Journal of Software
Engineering.
Shetty, Jyothi and Demian Antony D’Mello. 2015. “Quality
of Service Driven Cloud Service Ranking and Selection
Algorithm Using REMBRANDT Approach.” in 2015
International Conference on Smart Technologies and
Management for Computing, Communication,
Controls, Energy and Materials, ICSTM 2015 -
Proceedings.
Siegel, Jane and Jeff Perdue. 2012b. “Cloud Services
Measures for Global Use: The Service Measurement
Index (SMI).in Annual SRII Global Conference, SRII.
Somu, Nivethitha, Kannan Kirthivasan, and Shankar
Sriram Shankar. 2017. “A Computational Model for
Ranking Cloud Service Providers Using Hypergraph
Based Techniques.” Future Generation Computer
Systems 68:1430.
Sun, Le, Hai Dong, Farookh Khadeer Hussain, Omar
Khadeer Hussain, and Elizabeth Chang. 2014. “Cloud
Service Selection: State-of-the-Art and Future Research
Directions.” Journal of Network and Computer
Applications 45:13450.
Sundareswaran, Smitha, Anna Squicciarini, and Dan Lin.
2012. A Brokerage-Based Approach for Cloud Service
Selection.” Pp. 558–65 in Proceedings - 2012 IEEE 5th
International Conference on Cloud Computing,
CLOUD 2012. Honolulu: IEEE.
Verein Kompetenznetzwerk Trusted Cloud eV. 2016.
Criteria Catalogue for Cloud Services. Retrieved
(https://www.trusted-
cloud.de/sites/default/files/trusted_cloud_kriterienkata
log_v1_1_en.pdf).
Wang, Shangguang, Zhipiao Liu, Qibo Sun, Hua Zou, and
Fangchun Yang. 2014. “Towards an Accurate
Evaluation of Quality of Cloud Service in Service-
Oriented Cloud Computing.” Journal of Intelligent
Manufacturing 25(2):28391.
Whaiduzzaman, Md et al. 2014. “Cloud Service Selection
Using Multicriteria Decision Analysis.” The Scientific
World Journal 2014:10.
Wibowo, S. and H. Deng. 2016. “Evaluating the
Performance of Cloud Services: A Fuzzy Multicriteria
Group Decision Making Approach.” Pp. 327–32 in
2016 International Symposium on Computer,
Consumer and Control (IS3C).
Wirtz, Jochen and Christopher Lovelock. 2016. Services
Marketing. World Scientific Publishing Company.
Zheng, Zibin, Xinmiao Wu, Yilei Zhang, Michael R. Lyu,
and Jianmin Wang. 2013. “QoS Ranking Prediction for
Cloud Services.” IEEE Transactions on Parallel and
Distributed Systems 24(6):121322.
CLOSER 2019 - 9th International Conference on Cloud Computing and Services Science
248