ANALYSIS OF THE GRAPHICS PROCESSORS FOR MEDICAL PROBLEMS

Abstract

Abstract. The main function of graphics processors since their inception has
been graphics processing. Subsequently, after it became possible to program the
processing of model vertices and pixels of rendered three-dimensional scenes using
special programs (shaders), the architecture of graphic processors changed
significantly. After the advent of the first general-purpose programmable graphics
processors (G80 architecture from NVIDIA, R600 architecture from ATI), it became
possible to process commands not only for graphic data in vector form, but also to
perform ordinary calculations for arbitrary data on a variety of special cores, while
implementing data parallelism. Therefore, Graphics Processing Units (GPU) show
high efficiency rates when parallelizing programs that process a lot of data of the
same type. Such programs include shaders: a vertex shader processes 3D vertices
with different parameters, a pixel shader processes 2D pixels on the screen using
interpolated data. Even before the advent of general-purpose kernels, there were
attempts to simulate the processing of arbitrary data of the same type written into
textures (image matrices) using pixel shaders, which, of course, gave a performance
boost compared to the CPU.