Thesis Details
Knihovna pro profilování a vizualizaci spotřeby paměti programů C/C++
Performance is an important and, for some classes of programs, even critical aspect of user experience. The current approaches to performance profiling are, however, far from being satisfactory. The aim of this bachelor's thesis is to extend the current state-of-the-art of profiling and visualization solutions, with novel techniques which can be used for a more efficient search of performance bugs in C/C++ programs and their subsequent interpretation to the end user. Thesis briefly introduces existing tools dealing with similar problems and then proposes a novel solution to collection of profiling data and their illustrative interpretation. The resulting implementation is, moreover, integrated in the Perun --- Performance Control System --- platform for profile versioning. The functionality of the implementation is demonstrated on a series of non-trivial programs.
profiling, C/C++, visualization, memory, performance, Python, Bokeh, GCC
Janoušek Vladimír, doc. Ing., Ph.D. (DITS FIT BUT), člen
Novák Michal, doc. RNDr., Ph.D. (DMAT FEEC BUT), člen
Strnadel Josef, Ing., Ph.D. (DCSY FIT BUT), člen
Szőke Igor, Ing., Ph.D. (DCGM FIT BUT), člen
@bachelorsthesis{FITBT19095,
author = "Radim Podola",
type = "Bachelor's thesis",
title = "Knihovna pro profilov\'{a}n\'{i} a vizualizaci spot\v{r}eby pam\v{e}ti program\r{u} C/C++",
school = "Brno University of Technology, Faculty of Information Technology",
year = 2017,
location = "Brno, CZ",
language = "czech",
url = "https://www.fit.vut.cz/study/thesis/19095/"
}