Scientific research

Among many research and deployment activities of Distributed Systems Research Group I am involved in an exploration of component technologies in JAVA environment.

My disseration titlted "Scalable, component system for collecting and storing of data originating from monitoring of distributed systems" was tighty coupled with this area.

The aim of the dissertation was to prove the possibility of creation in a component technology of efficient and scalable system for collecting and storing data originating from the monitoring of distributed systems, and such representation of the data that would provide universality and efficiency of this solution. Growing demand for massive monitoring data storing is a real challenge for contemporary systems, mainly in terms of universality and scalability. In this context, universality is defined as an ability to collect and store any kind of monitoring data, as well as adaptability to data changes and support for different monitoring modes; scalability guarantees adequate system performance with increasing incoming volume of data stream to be stored. The dissertation presents a concept of meta-data for monitoring systems and universal object model for data originating from distributed systems monitoring. The research concerns possible solutions of component system architecture based on multiple database instances, data buffering, data partitioning and asynchronous communication. They are practically verified in complex environment with the use of proposed test sets and evaluation criteria in the context of achieved efficiency and scalability.

More information about the dissertation can be found in:

• Complete dissertation text (in polish),
• Dissertation summary (12 pages, in polish),
• Presentation given during dissertation defence (in polish).