CUBLAS-aided Long Vector Algorithms

Autor: D. L. Golovashkin, Darya Gennadievna Vorotnikova
Rok vydání: 2014
Předmět:
Zdroj: Journal of Mathematical Modelling and Algorithms in Operations Research. 13:425-431
ISSN: 2214-2495
2214-2487
DOI: 10.1007/s10852-014-9267-7
Popis: The paper propose vector methods that allow you to use GPU-processors more rationally. The approach is based on using long vectors of arguments instead of short matrix rows. Efficiency of the method is verified by comparisons with a library OpenCurrent.
Databáze: OpenAIRE
Externí odkaz: