Modeling and measurement of cloud services performance
Cloud services are emerging today as an innovative IT provisioning model, offering benefits over the traditional approach of provisioning infrastructure. However, the occurrence of multi-tenancy, virtualization and resource shar¬ing in the cloud raise certain difficulties in providing performance estimation during application design or deployment time. In order to assess the perfor¬mance of cloud services and compare cloud offerings from different cloud providers both the extension of an existing metamodel, namely CloudML@artist, for describing this information in a machine understandable format and cloud benchmarks are required. In this thesis context, both of these requirements have been implemented. Specifically, performance in¬stances for different cloud providers are implemented based on CloudML@artist metamodel and to complete the instance creation, the in¬corporation of a number of performance metric values in the concrete in¬stances for different cloud providers is provided. Performance measurement is achieved through benchmarking process and performance results are demonstrated from three large commercial cloud providers, Amazon EC2, Microsoft Azure and Flexiant, in order to support the provisioning decisions of the cloud users.