Analysis of automatic program parallelization based on bytecode There are many algorithms for automatic parallelization and this work explores the possible application of these algorithms to programs based on their bytecode or similar intermediate code. All these algorithms require the identification of independent code segments, because if two parts of code do not interfere with one another then they can be run in parallel without any danger of data corruption. Dependence testing is an extremely complicated problem and in general application, it is not algorithmically solvable. However, independences can be discovered in special cases and then they can be used as a basis for application of automatic parallelization, like the use of vector instructions. The first step is function inlining that allows the compiler to analyze the code more precisely, without unnecessary dependences caused by unknown functions. Next, it is necessary to identify all control flow constructs, like loops, and after that the compiler can attempt to locate dependences between the statements or instructions. Parallelization can be achieved only if the analysis discovered some independent parts in the code. This work is accompanied by an implementation of function inlining and code analysis for the .NET framework.

This diploma thesis is concerned with research on available methods of plagiarism detection and then with design and implementation of such detector. Primary aim is to detect plagiarism within academic works or theses issued at BUT. The detector uses sophisticated preprocessing algorithms to store documents in its own corpus (document database). Implemented comparison algorithms are designed for parallel execution on graphical processing units and they compare a single subject document against all other documents within the corpus in the shortest time possible, enabling near real-time detection while maintaining acceptable quality of output.

Cílem této práce je návrh nové techniky pro stabilizaci obrazu za pomoci hardwarové akcelerace prostřednictvím GPGPU. Využití této techniky umožnuje stabilizaci videosekvencí v reálném čase i pro video ve vysokém rozlišení. Toho je zapotřebí pro ulehčení dalšího zpracování v počítačovém vidění nebo v armádních aplikacích. Z důvodu existence vícerých programovacích modelů pro GPGPU je navrhnutý stabilizační algoritmus implementován ve třech nejpoužívanějších z nich. Jejich výkon a výsledky jsou následně porovnány a diskutovány.

Graphical Processing Units stand for the success of Artificial Neural Networks over the past decade and their broader application in the industry. Another promising field of Artificial Intelligence is Evolutionary Algorithms. Their parallelization ability is well known and has been successfully applied in practice. However, these attempts focused on multi-core and multi-machine parallelization rather than on the GPU. This work explores the possibilities of Evolutionary Algorithms parallelization on GPU. I propose implementation in PyTorch library, allowing to execute EA on both CPU and GPU. The proposed implementation provides the most common evolutionary operators for Genetic Algorithms, Real-Coded Evolutionary Algorithms, and Particle Swarm Op- timization Algorithms. Finally, I show the performance is an order of magnitude faster on GPU for medium and big-sized problems and populations. 1

Monitoring and security in multigigabit networks with speeds 1 - 100 Gb/s needs hardware acceleration. NetCOPE platform for rapid development of network applications uses hardware acceleration card with FPGA technology by means of hardware/software codesign. Increas in performance of platform's software part is dependent of parallel processing in applications to take advantage of utilising more processor cores. This thesis analyses NetCOPE platform architecture and possibilities of parallelising classic network applications and creates models of concurrent access to data in NetCOPE platform to utilize more processor cores. These models are subsequently implemented as extensions to platform's Linux system drivers. Userspace libraries are created to provide simple interface for applications to use these new features. To achieve high throughput of this solution several optimizations are performed. Results are measured by created testing tools.

In this thesis, a program for fracture test in civil engineering has been optimized. The simulation is used for a validation of the fracture characteristics for blocks of construct material used for historic buildings reconstructure. This thesis illustrates the possibilities of an effective usage of the processor’s potential without the loss of the output quality. The individual parts of the simulation are analyzed and this thesis proposes for the critical sections some possible optimizations such as vectorization or parallel processing. The techniques used in this thesis may be used on similar computing problems and help shorten the required runtime. The prototype of the simulation was able to process the simulation in 7.7 hours. Optimized version is capable to process the same simulation in 2.1 hours on one core or 21 minutes on eight cores. The parallel optimized version is 21 times faster than the prototype.

This thesis deals with the design and implementation of a system, which allows to search for and recognize people in video recordings. The presented design is based on a preceding research in theory relating to the topics of face and people recognition. Furthermore, the system design is implemented using convolutional neural networks for face recognition, while the implementation primarily utilizes the libraries dlib and OpenFace. The design and implementation use parallelization and distribution of tasks among multiple devices to reduce computation time, while also bearing in mind the practical applications of such system, such as working with limited amounts of available information regarding the person we seek. The precision of people detection and recognition of the implemented system is about 70% to 80%, based on the performed task. Among other uses, the system can be utilized to find a particular person in a video recording, to estimate the number of passes through the monitored space of one person, or the number of passes in total, or to find unknown people in the monitored space.

The master's thesis deals with possible ways of accelerating numerical computations, which are present in problems related to heat conduction in solids. The thesis summarizes basic characteristics of heat transfer phenomena with emphasis on heat conduction. Theoretical principles of control volume method are utilized to convert a direct heat conduction problem into a sparse linear system. Relevant fundamentals from the field of inverse heat conduction problems are presented with reference to intensive computations of direct problems of such kind. Numerical methods which are well-suited to find a solution of direct heat conduction problems are described. Remarks on practical implementation of time-efficient computations are made in relation with a two-dimensional heat conduction model. The results are compared and discussed with respect to obtained computational time for several tested methods.

Stochastické programování a optimalizace jsou mocnými nástroji pro řešení široké škály inženýrských problémů zahrnujících neurčitost. Algoritmus progressive hedging je efektivní dekompoziční metoda určená pro řešení scénářových stochastických úloh. Z důvodu vertikální dekompozice je možno tento algoritmus implementovat paralelně, čímž lze významně ušetřit výpočetní čas a ostatní prostředky. Teoretická část této diplomové práce se zabývá matematickým a zejména pak stochastickým programováním a detailně popisuje algoritmus progressive hedging. V praktické části je navržena a diskutována původní paralelní implementace algoritmu progressive hedging, která je pak otestována na jednoduchých úlohách. Dále je uvedená paralelní implementace použita pro řešení inženýrského problému plynulého odlévání ocelové bramy a na závěr jsou získané výsledky zhodnoceny.

Dizertační práce se zabývá algoritmy SegDWT pro segmentový výpočet Diskrétní Waveletové Transformace – DWT jedno i vícedimenzionálních dat. Segmentovým výpočtem se rozumí způsob výpočtu waveletové analýzy a syntézy po nezávislých segmentech (blocích) s určitým překryvem tak, že nevznikají blokové artefakty. Analyzující část algoritmu pracuje na principu odstranění přesahu a produkuje vždy část waveletových koeficientů z waveletové transformace celého signálu, které mohou být následně libovolně zpracovány a podrobeny zpětné transformaci. Rekonstruované segmenty jsou pak skládány podle principu přičtení přesahu. Algoritmus SegDWT, ze kterého tato práce vychází, není v současné podobně přímo použitelný pro vícerozměrné signály. Tato práce obsahuje několik jeho modifikací a následné zobecnění pro vícerozměrné signály pomocí principu separability. Kromě toho je v práci představen algoritmus SegLWT, který myšlenku SegDWT přenáší na výpočet waveletové transformace pomocí nekauzálních struktur filtrů typu lifting.