Modeling and Prediction of I/O Performance in Virtualized Environments

In this thesis, we present a novel performance modeling approach tailored to I/O performance prediction in virtualized environments. First we identify important performance-influencing factors. Capturing these factors, we then develop tailored I/O performance models using complementary modeling formalisms. To increase the practical applicability of these models, we finally combine the low-level I/O performance models with high-level software architecture models.