Decomposition approach to the construction of parallel algorithms for processing of two-dimensional data
A way of constructing parallel algorithms for processing two-dimensional data (two-dimensional convolution and correlation, two-dimensional discrete Fourier transform (DFT) and other structurally similar operations) is considered. The proposed path is characterized by the use of a decomposition approach and the formation of compositional forms designed to describe parallel computations. New possibilities are presented, which open when these forms are used to implement computations of two-dimensional operations and allow to increase the efficiency of parallel processing.
Keywords
декомпозиционный подход, параллельные алгоритмы, композиционные формы, модельное описаниеAuthors
| Name | Organization | |
| Klimova Olga V. | Institute of Engineering Science, Ural Branch of the Russian Academy of Sciences | ovk31@mail.ru |
References
Decomposition approach to the construction of parallel algorithms for processing of two-dimensional data | Vestnik Tomskogo gosudarstvennogo universiteta. Upravlenie, vychislitelnaja tehnika i informatika – Tomsk State University Journal of Control and Computer Science. 2020. № 52. DOI: 10.17223/19988605/52/14