An SLA-based Framework for Estimating Trustworthiness of a Cloud

Abstract

In cloud computing consumers often seek some assurance from cloud service providers (CSPs) that services will be provided according to consumers' requirements. The "service-level-agreement" (SLA) between a CSP and a consumer solves this issue to an extent and provides some level of assurance. However, SLAs alone do not solve the issue entirely as there is no unique rule to create an SLA. They vary in description, length, and types of information released. Therefore, consumers need a better way of estimating 'trustworthiness' of a cloud (CSP). In this work we propose a framework to alleviate the above issue. Our framework estimates trustworthiness of a cloud using a quantitative model of trust. We identified and formalized several parameters that can be extracted from SLA or retrieved during the sessions and are used to estimate trust.

A Review on Various Trust Models in Cloud Environment

“…Policy database stores all the policies which are involved in trust estimation. Session log archive stores the log files of transactions of all sessions between the customer and the provider [8]. This model supports both cloud provider and service consumer, in which the user can decide to prolong or suspend the services with the service provider.…”

Design and Comparative Analysis of MCDM-based Multi-dimensional Trust Evaluation Schemes for Determining Trustworthiness of Cloud Service Providers

“…Chakraborty et al [16] propose a quantitative model to estimate trustworthiness of a cloud. To estimate trust, parameters are extracted from SLA and similar documents, which are then processed to compute trustworthiness as a fraction between 0 and 1.…”

“…To the best of our knowledge, no comprehensive comparison of MCDM-based trust evaluation system exists in literature which evaluates trustworthiness of a CSP based on its compliance to SLA, except for some attempts [14][15][16][17]. Though there are other similar works [18][19][20][21][22][23], which also evaluate trustworthiness of a CSP, but they do not establish trustworthiness based on compliance of a CSP to SLA, which, we argue, is very necessary to truly label a CSP as trustworthy or untrustworthy as per trustworthiness definition defined earlier.…”

Security Isolation Strategy Mechanism for Lightweight Virtualization Environment

“…Technology needed by second assurance requirements is relatively a bit more complicated, such as through the precise formal verification [4]to describe behavior standards, such as ensure that the source ( developed in-house or a third party )of the specification is transparent controllable [5][6][7]: if from developed in-house , the user can through process management, code size control methods to minimize or eliminate potential errors or bugs in code of conduct; if from a third party, the user can through other ways to evaluate the credibility of the code of conduct [8][9][10], such as the evaluation of others, such as solve the operation control issues of the user program with reference monitor [11,12]. System credible guarantee are shown in Figure 1 Figure 1.…”