Tools for parallel computing in computer graphic tasks

Authors

  • Д. П. Кучеров
  • К. О. Моргун
  • Л. С. Онікієнко

DOI:

https://doi.org/10.18372/2310-5461.38.12833

Keywords:

parallel computing, programming tools, computing tools, computer graphics, Monte Carlo method

Abstract

The paper considers the means of ensuring acceleration of computations that arise when working with large volumes of data. It has been established that in addition to computer graphics, such a problem arises in problems not only for problems of linear algebra, where the data are structured, but also with unstructured data. It is noted that the natural approach to solving this problem is to use parallel computations, which can be performed in the multiprocessor architecture of computers and parallel programming. Known approaches based on hardware implementation to address specific tasks are costly from resources and, within a separate organizational structure, are not suitable for solving settlement tasks of different types. At the same time, other approaches are being developed to support well-known processor structures and existing software for them. The approach based on the application of a parallel structure is proposed in the paper for solving the rendering problem, which is based on the integral equation of illumination by the Monte Carlo method. The peculiarity of solving this problem is the transition from the general equation to the three equations corresponding to the adopted color model of the image, namely the RGB model. The solution is in the spatial coordinates of color, which requires a further transition from spatial coordinates to real coordinates of color. Apart from, in the final decision, the correction of the nonlinearity of the display device is additionally performed. To apply parallel calculations to solve this problem in the work, it is proposed to partition the original image into parts in accordance with the number of processors that participate in the calculations and ensure their synchronization with appropriate software management tools. A number of examples of the Monte Carlo method for solving the above mathematical problems are presented, among them the integration of the function and the development of this approach to work with images. The results of the solution of this problem are also given during the creation of a photorealistic image, as well as the analysis of the Amdahl’s formula given solution.

References

Воеводин В.В. Параллельные вычисления / Воеводин В.В., Воеводин Вл.В. – СПб.: БХВ-Петербург, 2002. – 608 с.

Параллельные вычисления [Электронный ресурс] – Режим доступа: https://ru.wikipedia.org/wiki/Параллельные_вычисления.

Введение в параллельные вычисления [Электронный ресурс] – Режим доступа: http://habrahabr.ru/post/126930.

Долинский М. Обзор аппаратных и программных средств реализации параллельной обработки / М. Долинский, А. Толкачев // Компоненты и технологии. - № 4. – 2006. [Электронный ресурс] – Режим доступа: http://www.kit-e.ru/assets/files/pdf/2004_06_152.pdf

Моргун К. О. 3D моделювання анімованих персонажей / К. О. Моргун, Д.П. Кучеров // Наукоємні технології. – № 2 (26). – 2015. – С. 133-140.

Kucherov D.P. А visualization of 3D images by Back Ray Tracing: the Geometric Approach / D.P. Kucherov, К.О. Morgun, T.I. Golenkovskaya // Electonic and Control systems. – No. 2(48). – 2016. – p. 140-144.

Morgun K. O. Calculation of integrals by Monte Carlo in the illumination problem of synthesized objects / D. P. Kucherov, I. V. Ogirko, O. I. Ogirko, K. O. Morgun, T. I. Golenkovskaya // Science and education a new dimension. Natural and Technical Sciences. Vol. IV (11). – Is. 96. – 2016. – Р.42-47.

Моргун К. О. Інформаційна технологія рендерінгу 3D-обєктів / К. О. Моргун // Наукоємні технології. – № 3 (35). – 2017. – С. 189 – 196, doi: 10.18372/2310-5461.35.11837 (укр)

Issue

Section

Information and Communication Systems and Networks