×

You are using an outdated browser Internet Explorer. It does not support some functions of the site.

Recommend that you install one of the following browsers: Firefox, Opera or Chrome.

Contacts:

+7 961 270-60-01
ivdon3@bk.ru

Problems of evaluating the effectiveness of distributed and multiprocessor systems of various classes for image processing tasks

Abstract

Problems of evaluating the effectiveness of distributed and multiprocessor systems of various classes for image processing tasks

Kostyuk A.I., Pukhovsky V.N., Gorbunov F.V., Korobeynikova N.M.

Incoming article date: 13.06.2019

A comparative analysis of the use of distributed and multiprocessor systems of various classes for image processing tasks is considered. Designations and classifications of calculations, the nature of frequently used operations for image processing, the classification of architectures for image processing are given. Issues of correspondence of general classes of computational architectures to types of computational operations used in image processing are investigated. It is shown that each architecture has advantages and disadvantages in some special processing areas, but none of them is optimal for all cases. The most appropriate approach to determining the performance of distributed and multiprocessor systems in this case is the process of their architectural modeling, followed by emulation on the resulting HINT-type test model. The advantages of this approach are the possibility of obtaining a reliable estimate of the performance of the entire system as a whole, as well as the simplicity of its implementation for various architectures.

Keywords: image processing algorithms, distributed systems, multiprocessor computing systems, image analysis, architecture classification