The complex IT infrastructures of today are becoming more and more of a challenge to the engineers overseeing, controlling, and maintaining them. At the same time demands for reliability and availability are increasing, further complicating their task.
My work is about the reliability modeling of complex IT infrastructures, in which my aim was to aid the model based inspection of these complex systems.
In my thesis I introduce tools with the help of which a complex IT infrastructure can be mapped, the structure’s model can be created, and can be put under further analysis. I primarily used academic tools because of their free licensing terms.
To automate the examinations I’ve developed an application that supports conversion from the examined system’s intermediate model to stochastic Petri nets, which it can save in the standard PNML format. We can use Petri nets created by this process to conduct reliability evaluations on them with the help of analysis tools.
At the end of my thesis I demonstrate the usage of the developed modeling conventions using smaller models. With these same models I also demonstrate the possibilities in the modeling and evaluation of SLA criteria, and provide measurements to prove worthiness of this approach.