Generating Trustworthiness Adaptation Plans Based on Quality Models for Cloud Platforms

• Jorge Luiz Machado da Silva UNICAMP
• Breno B. Nicolau de França UNICAMP
• Cecília Mary Fischer Rubira UNICAMP

Resumo

Cloud computing platforms can offer many benefits related to the provision of service processing and storage for hosting client applications. Trustworthiness can be defined as the trust of a customer in a cloud service and its provider; however, the assurance of this property is not trivial. First, trustworthiness in general is not composed by a single quality attribute, but by the combination of multiple attributes, such as data privacy, performance, reliability, etc. Second, during runtime clients can experience a change of the trustworthiness level required by their application due to the degradation of the cloud service. This article presents a solution that monitors during runtime the set of quality attributes of a specific application and generates adaptation plans in order to certify that an adequate resource amount be provided by the cloud in order to keep its trustworthiness level. Our solution is based on quality models to compute the metric associated to each non-functional requirement and their combination them into different types of trustworthiness levels. The main contribution of the solution is to provide an approach which deals with multiple requirements at the same time (or simultaneously) during runtime in order to adapt the cloud resources to keep the trustworthiness level required by the application. The solution was evaluated by an experiment considering a scenario where the application trustworthiness level was composed by three quality attributes: data privacy, performance and reliability. Initial results have shown that the approach is feasible in terms of the execution of the adaptation plans during runtime to certify the trustworthiness level required by the application.

Palavras-chave: Adaptation Planning, Cloud Computing, Self-adaptive Systems, Trustworthiness