PRESEnCE: A Framework for Monitoring, Modelling and Evaluating the Performance of Cloud SaaS Web Services

摘要

Cloud Services Providers (CSPs) provide cloud services to the cloud customers in the pay-per-use model and use the Service Level Agreements (SLAs) to define the quality of the provided services. SLAs are just a contract which characterizing the performance and quality of the CSPs' services. Unfortunately, there is not an automatic and standard mechanism to verify and assure that delivered services satisfy the signed SLA agreement. In this context, this work aim at developing an automatic framework called PRESENCE, to evaluate the Quality of Service (QoS) for the deployment of the Software-as-a-Service (SaaS) Web Services (WSs) offered across several CSPs. This performance evaluation will be used to verify and check on the SLA contraventions. PRESENCE aims at quantifying in a fair and by stealth way the performance and scalability of the delivered WS. By stealthiness, we refer to the capacity of evaluating a given Cloud service through multiple workload patterns that makes them indistinguishable from a regular user traffic from the provider point of view. PRESENCE introduces a definition for the set of Common performance metrics for measuring the behaviour of cloud applications on top of a given CSP. This framework opens a novel perspectives for SLA contraventions assessment, monitoring, modelling the performance metrics for SaaS WSs.