Comprehensive Framework for Selecting Cloud Service Providers (CSPs) Using Meta synthesis Approach

Nowadays, cloud computing has attracted the attention of many organizations. So many of them tend to make their business more agile by using flexible cloud services. Currently, the number of cloud service providers is increasing. In this regard, choosing the most suitable cloud service provider based on the criteria according to the conditions of the service consumer will be considered one of the most important challenges. Relying on previous studies and using a meta-synthesis approach, this research comprehensively searches past researches and provides a comprehensive framework of factors affecting the choice of cloud service providers including 4 main categories and 10 sub-areas. Then, using the opinions of experts who were selected purposefully and using the snowball method, and using the Lawshe validation method, the framework is finalized.

This research aims to complete the results of previous studies and answer the following questions with a systematic review of the subject literature:-What are the components of the comprehensive framework for choosing cloud service providers?
-What are the effective criteria to choose a cloud service provider?
-What is the selected framework of effective factors?

Many researchers have looked at the problem of choosing the best CSP from different aspects and have tried to provide a solution in this field. In this regard, we can refer to "Tang and Liu" (2015) who proposed a model called "FAGI" which defines the choice of a trusted CSP through four dimensions: security functions, auditability, management capability, and Interactivity helps. "Kong et al." (2013) presented an optimization algorithm based on graph theory to facilitate CSP selection. Some researchers have also provided a framework for CSP selection, such as "Gash" (2015) who provides a framework called "SelCSP" with the combination of trustworthiness and competence to estimate the risk of interaction. "Brendvall and Vidyarthi" (2014) suggest that in order to choose the best cloud service provider, a customer must first identify the indicators related to the level of service quality related to him and then evaluate different providers. Some researchers have focused on using different techniques for selection. For example: "Supraya et al." (2016) use the MCDM method to rank based on infrastructure parameters (agility, financial, efficiency, security, and ease of use). They investigate the mechanisms of cloud service recommender systems and divide them into four main categories and their techniques in four features of scalability, accessibility, accuracy, and trust
In this research, it has been tried to use the models and variables of the subject literature in developing a comprehensive framework. The codes, concepts, and categories related to the choice of cloud service providers are extracted from previous studies, and a comprehensive framework of the factors influencing the choice of cloud service providers is presented using the meta-composite method.

In this research, based on the "Sandusky and Barroso" meta-composite qualitative research method, which is more general, a systematic review of the research literature was conducted, and the codes in the research literature were extracted. Then the codes, categories, and finally the proposed model are formed. The seven-step method of "Sandusky and Barroso" consists of: formulation of the research question, systematic review of the subject literature, search and selection of suitable articles, extraction of article information, analysis and synthesis of qualitative findings, quality control, and presentation of findings. Lawshe validation method has been used to validate the research findings.

In the meta-synthesis method, all the factors extracted from previous studies are considered as codes and concepts are obtained from the collection of these codes. Using the opinion of experts and considering the concept of each of these codes, codes with similar concepts were placed next to each other and new concepts were formed. This procedure was repeated in converting the concepts into categories and the proposed framework was identified. This framework consists of 27 codes, 10 concepts, and 4 categories (Table 1).
Table 1: Codes, concepts, and categories extracted from the sources
category
Concept
Code
No.
Trust
Security
Hardware Security
1 Network Security
2 Software Security
3 Confidentiality
4 Control
5 Guarantee and Assurance
Accessibility
6 Stability
7 Facing Threats
Technical Risk
8 Center for Security Measures
9 Technology
Efficiency
Service Delivery Efficiency
10 Interactivity
11 Hardware and Network Infrastructure
Configuration and Change
12 Capacity (Memory, CPU, Disk)
13 Functionality
Flexibility
14 Usability
15 Accuracy
16 Service Response Time
17 Ease of use
18 Managerial
Maintenance
Education and Awareness
19 Customer Communication Channels
20 Strategic
Legal Issues
21 Data Analysis
22 Service Level Agreement
23 Commercial
Customer Satisfaction
Responsiveness
24 Customer Feedback
25 Cost
Subscription Fee
26Implementation Cost
27The lack of a common framework for evaluating cloud service providers is compounded by the fact that no two providers are the same, so that this issue complicates the process of choosing the right provider for each organization. Figure 1 shows the proposed comprehensive framework including 4 categories and 10 concepts covering the issue of choosing cloud service providers. These factors are useful in determining the provider that best matches the personal and organizational needs of the service recipient. The main categories are: trust building, technology, management, and business, which will be explained in the following.
Figure 1: Cloud service provider selection framework

By comprehensively examining the factors affecting the choice, this research introduces specific areas such as trust building, technology, management, and business as the main areas of cloud service provider selection and add to the previous areas. The category of building trust between the customer, and the cloud service provider is of particular importance. In this research, the concepts related to trust building are: security (including hardware security, network security, software security, confidentiality and control), (availability, stability and stability), and facing threats (technical risk). In 36% of the articles, the concept of trust is mentioned, but in each study, only a limited number of factors affecting this category are discussed. This research takes a comprehensive look at the category of technology, the concepts of productivity (including service delivery efficiency, interactivity), hardware and network infrastructure (including configuration and repair, capacity (memory, processor, disk)), and performance (including flexibility, usability, accuracy of operation, service response time, ease of use). Considering the variety of services on different cloud platforms, service recipients must ensure that the provision of services is managed easily and in the shortest possible time by the cloud provider. The commercial aspect of service delivery deals with the two concepts of customer satisfaction (including responsiveness, customer feedback) and service rates (including: subscription cost and implementation cost), which are of interest to many businesses. The results of this research will help the decision makers of using the cloud space (both organizational managers and cloud customers) in choosing the best cloud service provider to have a comprehensive view of the effective factors before choosing and plan according to their needs.