Abstract
Most software systems need to be adapted during their life cycle. It is estimated that more than 60% of the cost of a software system is related to its maintenance and evolution. The evolution of a software system not only involves the modification made to its various components, but it also includes an indispensable preliminary phase, which is the in-depth understanding of each component of the system. This is especially true for a certain type of systems called data-intensive systems. Within such a system, the interactions between the application programs and the database are becoming increasingly difficult to analyze, and therefore to understand. This is why program understanding in general has become an important topic of interest for software engineering researchers and developers. In particular, understanding today’s data-intensive systems clearly calls for automated support.The goal of this thesis is to facilitate the understanding of large data intensive systems using dynamic analysis, visualization and process mining techniques. The dynamic analysis techniques seek to analyze and visualize the data-manipulation behavior of data-intensive systems via the analysis of their SQL execution traces. The process mining techniques seek to retrieve recurring patterns and extract the data-manipulation processes followed by the program itself.
Date of Award | 19 Jun 2018 |
---|---|
Original language | English |
Awarding Institution |
|
Supervisor | Anthony Cleve (Supervisor), Naji Habra (Co-Supervisor), Wim Vanhoof (President), Vincent Englebert (Jury), Tom Mens (Jury) & Alexander Serebrenik (Jury) |