Parallel scalable parsing with Floyd grammar

Recent development in the hardware sector has made parallelized architectures almost pervasive. While parallel solutions have been found for many common problems, this is not the case for parsing, where traditional techniques are inherently sequential, except for specific ad-hoc solutions. From recent studies in the field of formal languages, and in particular Floyd grammars, the possibility for a parallel scalable language-independent algorithm has surfaced. In this thesis the study of the problem is dealt with and a possible solution is presented. As a proof of concept the development of the algorithm and experimental results comparing it with traditional sequential solutions accompany this work. This thesis is part of a Research Project that won a Google Research Award.