Cybernetics and programming
Journal Menu
> Issues > Rubrics > About journal > Authors > About the Journal > Requirements for publication > Council of Editors > Peer-review process > Policy of publication. Aims & Scope. > Article retraction > Ethics > Copyright & Licensing Policy > Digital archiving policy > Open Access Policy > Open access publishing costs > Article Identification Policy > Plagiarism check policy
Journals in science databases
About the Journal

MAIN PAGE > Journal "Cybernetics and programming" > Rubric "Parallel algorithms for numerical analysis"
Parallel algorithms for numerical analysis
Sechenov P., Olennikov A.A. - Application of NVIDIA CUDA parallel programming technology in the task of melting a spherical particle pp. 8-14


Abstract: The article describes the NVIDIA CUDA parallel programming technology used  in the task of melting a spherical particle. The tendency of modern computers to increase power by increasing the number of cores, and not by increasing the frequency of the processor (which leads to significant energy consumption and heat generation). The Amdal law is presented, allowing to estimate the acceleration of the program time when parallelized on N processors. The conditions for increasing the performance of the algorithm in parallelizing tasks are listed. The task of melting iron ore particles is presented. The features of the parallel programming language CUDA C are considered and the algorithms for the selected task are presented. A comparative analysis of the task execution time on the CPU (C #) and GPU (CUDA C) has been made. The technology of parallel programming CUDA allows you to increase the performance of parallelized algorithms of complexity N up to 60 times. This requires the presence of a graphics processor supporting this technology, the development environment and the CUDA compiler, knowledge of the CUDA C language, as well as a good knowledge of the task and the possibility of its parallelization.
Pekunov V.V. - Application of prediction in parallel processing of chains of predicates in regular-logic expressions pp. 48-55


Abstract: This paper addresses the problem of choosing the execution mode (sequential or parallel) when processing chains of predicates in regular-logic expressions. A brief description of the essence of regular-logical expressions, their known applications (natural language interfaces, automatic parallelizer of C-programs), types and composition of predicate chains is given. Particular attention is paid to the question of the prediction of time spent on processing chains in one mode or another. Various approaches to such a possible prediction are considered in detail. It is noted that in this case the semi-empirical-statistical approach is the most natural. The paper uses the basic relations of the theory of parallel computing, interpolation and extrapolation methods, computational experiment, elements of statistical processing. A new semi-empirical-statistical approach to solving the problem of calculating estimates of the execution time of chains of predicates is proposed. The approach is distinguished by the minimum amount of time measurement achieved using partial recovery of missing data, and the use of potentially more accurate linear autoregressive and quadratic models to calculate the estimated execution time in sequential or parallel modes.
Other our sites:
Official Website of NOTA BENE / Aurora Group s.r.o.
"History Illustrated" Website