Artificial intelligence

Scientific journal

ISSN 2710-1673

ONLINE: ISSN 2710-1681

Search by:

Year of publication
Author name
Paper title

Parallel addition based on vector operations

Tereshchenko A.1, Zadiraka V.1
1 VM Glushkov Institute of Cybernetics of NAS of Ukraine

Full text (PDF)

UDC: 519.6
Publication Language: Russian
Stuc. intelekt. 2018; 23(2):122-137

Abstract: The article describes the new method of implementation of multidigit addition in parallel model of computation. The method uses vector operations that reduces noticeably the number of stream processors. The article describes algorithm in parallel model of computation. The analysis of complexity of described algorithm shows that in parallel model of computation the number of one-digit operations 4к+59 depends linearly on number of processors k used in calculation taking into account that every processor executes constant number of vector operations of the length of 16. Algorithm is implemented using language OpenCL (v. 1.2) and tested.

Keywords:

References:

  1. Catherine C. McGeoch, Parallel Addition, TheAmerican Mathematical Monthly, Vol. 100, No. 9(Nov., 1993), pp. 867-871.(http://www.jstor.org/stable/2324666).
  2. T. H. Cormen, C. E. Leiserson, and R. L. Rivest,Introduction to Algorithms, MIT Press 1990.
  3. Tereshchenko A.N., Zadiraka V.K. Otsenkaslozhnosti operatsii umnozheniyamnogorazryadnyih chisel v parallelnoy modelivyichisleniy // Kompyuternaya matematika. –2016. – # 1.
  4. Tereshchenko A.N. Analiz skladnosti operatsiimnozhennia bahatorozriadnykh chysel pryrealizatsii u paralelnii modeli obchyslen //Problemy prohramuvannia. – 2015. – № 1.
  5. Tereshchenko A.M. Optymizatsiiabahatorozriadnoho mnozennia na osnovi ShPF uparalelnii modeli obchyslen // Zakhyst informatsii.– 2014. – № 3. – S.178–184.RESUME

View full text (PDF)