A fuzzy inference system (FIS) to evaluate the security readiness of cloud service providers

Syed Rizvi,John Mitchell,Iyonna Williams,Mohammad R Rizvi,Abdul Razaque

doi:10.1186/s13677-020-00192-9

Abstract

Cloud computing is a model for on-demand delivery of IT resources (e.g., servers, storage, databases, etc.) over the Internet with pay-as-you-go pricing. Although it provides numerous benefits to cloud service users (CSUs) such as flexibility, elasticity, scalability, and economies of scale, there is a large trust deficit between CSUs and cloud service providers (CSPs) that prevents the widespread adoption of this computing paradigm. While some businesses have slowly started adopting cloud computing with careful considerations, others are still reluctant to migrate toward it due to several data security and privacy issues. Therefore, the creation of a trust model that can evolve to reflect the true assessment of CSPs in terms of either a positive or a negative reputation as well as quantify trust level is of utmost importance to establish trust between CSUs and CSPs. In this paper, we propose a fuzzy-logic based approach that allows the CSUs to determine the most trustworthy CSPs. Specifically, we develop inference rules that will be applied in the fuzzy inference system (FIS) to provide a quantitative security index to the CSUs. One of the main advantages of the FIS is that it considers the uncertainties and ambiguities associated with measuring trust. Moreover, our proposed fuzzy-logic based trust model is not limited to the CSUs as it can be used by the CSPs to promote their services through self-evaluation. To demonstrate the effectiveness of our proposed fuzzy-based trust model, we present case studies where several CSPs are evaluated and ranked based on the security index.

Highlights

Cloud computing is a distributed computing environment capable of storing large quantities of data, increasing data processing efficiency, and scaling an application based on demand [1]
Since the input of the cloud service users (CSUs) are the source of our data sets, the use of subjective trust can be well justified with our fuzzylogic based trust model
We analyze the performance of the proposed fuzzy logic for cloud service providers (CSPs) using the fuzzy inference system to show that the CSPs are trustworthy

Summary

Introduction

Cloud computing is a distributed computing environment capable of storing large quantities of data, increasing data processing efficiency, and scaling an application based on demand [1]. Our use of fuzzy logic addresses the emotional aspect of subjective trust as it analyzes the uncertainties that may arise when users assess the reliability of a CSP as well as their inclination to trust a certain service because of aesthetics. Use of Fuzzy logic to evaluate CSPs with a new trust model: Our first main contribution is utilizing the process of fuzzification and defuzzification on the set of input data in order to evaluate service providers. Our contribution differs from other research as we use the fuzzy logic process for CSP security assessment and to establish trust between a CSU and CSP even in the presence of impartial data sets

Developing experimental case studies

Trust level validation of a CSP

Results and discussion

Conclusion