The general subject of my interest is finding concurrency bugs in complex software systems written in object-oriented programming languages. Concurrent, or multi-threaded, programming has become very popular in recent years. However, as concurrent programming is far more demanding than sequential, its increased use leads to a significantly increased number of bugs that appear in commercial software due to errors in synchronization. This stimulates a more intensive research in the field of detecting of such bugs. Despite persistent effort of wide community of researchers, a satisfiable solution of this problem for common programming languages like Java does not exist. I have focused on combination of three already existing approaches: (i) dynamic analysis which is in certain cases able to precisely detect bugs along an execution path, (ii) static analysis which is able to collect various information concerning tested application, and (iii) systematic testing that helps to examine as many different execution paths as possible. Moreover, I plan to incorporate artificial intelligence algorithms into process of testing of complex concurrent software and for bugs that are hard to detect I consider development of self-healing methods that can suppress manifestation of detected bugs during execution. | 
Fakulta informačních technologií
![[dnssec]](/common/img/dnssec.png)