|
|
Efficient Implementation of High Performance Algorithms on Intel Xeon Phi
Šimek, Dominik ; Hrbáček, Radek (referee) ; Jaroš, Jiří (advisor)
This thesis is dedicated to the implementation of high performance algorithms on the Intel Xeon Phi coprocessor. The Xeon phi was introduced by Intel as a new MIC (Many Integrated Core) architecture in 2012. The theoretical part of the thesis is focused on the architecture of the coprocessor (with peak performance of 2 tFLOPS for a single precision data) and on the procedure of algorithms implementation and optimization. The theoretical knowledge is then applied to a practical examples with demonstration of the implementation and the optimization of algorithms and work with the coprocessor. In the practical part of the thesis, simple benchmarks such as a vector matrix multiplication and a matrix multiplication are explained and implemented. In the first benchmark 6.5% of theoretical coprocessor performance was achieved, in the second it was much more. In following chapter a more complex benchmark - simulation of a particles system (N-Body), that reached more than 35% of coprocessor performance (725 gFLOPS), is discussed. The following section is dedicated to some interesting problems such as optimization of a MATLAB module k-Wave (propagation of the ultrasound waves), extraction of I-vector (speech processing), cross-compilation of existing libraries, modules and programs. In the conclusion of the thesis the usage the potential of the Intel Xeon Phi is evaluated.
|
|
| |
|
|
Fast Reconstruction of Photoacoustic Images
Kukliš, Filip ; Dvořák, Václav (referee) ; Jaroš, Jiří (advisor)
The ability of reconstruction of photoacoustic images is important requirement to study soft tissues or vascular and lymphatic systems in high resulution but in small space. Today solution needs extensive computing power and it is noticeably time-consuming. In this study we would like to introduce a new solution which would be a way much faster and easy to use. My solution is up to 20x faster and needs forty percent less memory. This solution may be a better alternative for scietnist who study soft tissues by photoacoustic imaging.
|
|
|
Parallelisation of Ultrasound Simulations Using Local Fourier Decomposition
Dohnal, Matěj ; Hrbáček, Radek (referee) ; Jaroš, Jiří (advisor)
This document introduces a brand new method of the 1D, 2D and 3D decomposition with the use of local Fourier basis, its implementation and comparison with the currently used global 1D domain decomposition. The new method was designed, implemented and tested primarily for future use in the simulation software called The k-Wave toolbox, but it can be applied in many other spectral methods. Compared to the global 1D domain decomposition, the Local Fourier decomposition is up to 3 times faster and more efficient thanks to lower inter-process communication, however it is a little inaccurate. The final part of the thesis discusses the limitations of the new method and also introduces best practices to use 3D Local Fourier decomposition to achieve both more speed and accuracy.
|
|
|
The GPU Based Acceleration of Neural Networks
Šimíček, Ondřej ; Jaroš, Jiří (referee) ; Petrlík, Jiří (advisor)
The thesis deals with the acceleration of backpropagation neural networks using graphics chips. To solve this problem it was used the OpenCL technology that allows work with graphics chips from different manufacturers. The main goal was to accelerate the time-consuming learning process and classification process. The acceleration was achieved by training a large amount of neural networks simultaneously. The speed gain was used to find the best settings and topology of neural network for a given task using genetic algorithm.
|
|
|
System for Advanced Scheduling
Horký, Aleš ; Jaroš, Jiří (referee) ; Drahošová, Michaela (advisor)
This master thesis deals with the automatic design of examinations and courses scheduling. The design is adapted to the specific requirements of the Faculty of Information Technology of Brno University of Technology. A genetic algorithm and a heuristic algorithm are employed to solve this task. The genetic algorithm is used to specify the sequence of the examinations (or the courses) and then the heuristic algorithm spread them out into a timetable. An implementation (written in Python 3) provides a fast parallel processing calculation which can generate satisfactory schedules in tens of minutes. Performed experiments show approximately 13% better results in all considered criteria in comparison with utilized examination schedules in the past. The development was periodically consulted with persons responsible for the schedule processing at the faculty. The program will be used while designing of examination schedules for the academic year 2015/2016.
|
|
|
Cost Management in the Company
Yablonskyy, Yevhen ; Ing.Jiří Jaroš (referee) ; Kocmanová, Alena (advisor)
The diploma thesis is focused on costs, calculation and primarily analysis of calculation systems and formula. The thesis is divided into two main parts – theoretical and practical part. The theoretical part is focused on explanation between financial and management accounting, description of basic terms and expense classification. Attention is given to system of calculation, its items, and calculation formula. Mention belongs to modern access of calculation – ABC. Practical part is focused on company Prefa Brno a.s., which main part of business is production and sale of building products and components. Shortly is mentioned characteristic of company, its history, present, sales and SWOT analysis. The main orientation of diploma thesis is on calculation system of the company, its descrtiption, methods of set-up and anylysis. Closing part of thesis includes summary of basic pieces of knowledge from analysis of calculation formula (system), which is compared with method of target casting, which belongs to modern methods of strategic management accounting. In the last part is given calculation in the case of additional item of material rate and quantification of impacts.
|
|
|
Customer Satisfaction Analysis and Proposals for its Improvement
Endrle, Pavel ; Kostková, Věra (referee) ; Jaroš, Jiří (advisor)
This master's thesis analyzes customer satisfaction of specific company and defines the methods and recommendations for improvement based on this analysis. Theoretical part is aimed at describing methods used in the analysis of satisfaction, which are then used for main research in the practical part of thesis. Methods and actions designed to increase customer's satisfaction based on main analysis of described company are proposed in practical part of thesis.
|
|
|
Evulation of the Financial Situation in the Firm and Proposals to its Improvement
Chalásová, Kristína ; Cesnak, Marek (referee) ; Jaroš, Jiří (advisor)
The thesis ‘Evaluation of the financial situation in the firm and proposals to its improvement’ deals with the financial analysis and evaluates the financial situation of the company. The first part defines the theoretical foundations of processing the financial analysis, its tools, methods and ways of using. The subject of the second part is the application of selected methods of financial analysis and proposed recommendations how to solve problem areas, reduce the weaknesses and improve the financial situation of the company.
|
|
| |
guest ::
RSS feed.