Optimal processing of requests in microservice based systems

OData support
Supervisor:
Dr. Majzik István
Department of Measurement and Information Systems

μSOA (Micro Service Oriented Architecture) systems are ontology driven architectures where every phase of the development and the execution is driven by a central ontology. The service catalog stores information about services (e.g. which abstract service is implemented over a specific content type and semantics, what is the name of the service implementation object and where it resides).

The Micro Service Server searches to satisfy the request for providers from the service catalog using dynamic pattern matching based on the abstract service and the concrete arguments in the request. Among them, one passes the request to the service provider that executes and then sends the response back to the client via the request generating micro-server service.

Since in μSOA systems service implementation for serving requests is selected dynamically, based on semantic information, existing SOA and web application server components are not directly usable.

In my work I present the requirements of a µSOA system, then I construct a component-based model te meet µSOA requirements. I examine the implementation options for each component, and assess them in the μSOA systems in terms of their applicability. Subsequently I examine the comparison options for testing the behavior of abstract service models and service models. Theoretical inquiries of different behavioral models were implemented and analytical measurements were performed on the model with the LQN tool. Measurement results are aggregated, and analyzed. Implications of the optimal μSOA request server are based on the measurement results.

Downloads

Please sign in to download the files of this thesis.