UDC 004.424.5.032.24
PARALLEL DATA SEARCH ALGORITHMS AND THEIR IMPLEMENTATION ON THE CUDA PLATFORM
S. V. Skvortsov, PhD (technical sciences), full professor, RSREU, Ryazan; This email address is being protected from spambots. You need JavaScript enabled to view it.
T. A. Fetisova, post-graduate student, RSREU, Ryazan; This email address is being protected from spambots. You need JavaScript enabled to view it.
D. V. Fetisov, post-graduate student, RSREU, Ryazan; This email address is being protected from spambots. You need JavaScript enabled to view it.
Theoretical and practical issues of parallel data search algorithms and their implementation on CUDA
platform are considered. The relevance of this topic is caused by constant increase in the volume
of processed information at present time, as well as the demand for increased speed of data processing.
The aim of the work is to modify the known algorithms for finding data in one-dimensional array for
their use on CPU, followed by the analysis of the efficiency of parallel versions of software modules developed.
Thus, the results show that the use of graphics accelerator can significantly increase the speed and,
accordingly, the acceleration of applications using data search algorithms. As the amount of data and thread
blocks used increases, the performance time of multithreaded application on graphics accelerator is significantly
reduced compared to serial implementation on CPU.
Key words: multiprocessor, device, graphic memory, kernel, block, thread, host, search, acceleration.