|
Numerical methods and programming, 2014, Volume 15, Issue 4, Pages 631–643
(Mi vmp279)
|
|
|
|
This article is cited in 1 scientific paper (total in 1 paper)
A parallel-processing-oriented method for the representation of multi-digit floating-point numbers
K. S. Isupov, A. N. Maltsev Vyatka State University
Abstract:
The extended precision of calculations is required in solving many scientific and engineering problems. The solution time is a critical parameter to accomplish and, therefore, new methods should be developed for fast high-precision arithmetic. In this paper a new modular-positional format for the representation of floating-point multi-digit numbers is proposed. The main concept of this format is to represent and ensure the digit-parallel processing of floating-point mantissas in residue number systems. The method of interval-positional characteristics is used to increase the speed of complex non-modular operations. Several algorithms for performing arithmetic operations and rounding in the new modular-positional floating-point format are considered. The results of studies of their vectorization efficiency and performance compared to some analogs (MPFR - Multiple Precision Floating-Point Reliable library, NTL - Number Theory Library, and Wolfram Mathematica) are discussed.
Keywords:
residue number system, high-precision computations, modular-position floating-point format, multi-digit numbers, arithmetic operations, high performance.
Received: 09.10.2014
Citation:
K. S. Isupov, A. N. Maltsev, “A parallel-processing-oriented method for the representation of multi-digit floating-point numbers”, Num. Meth. Prog., 15:4 (2014), 631–643
Linking options:
https://www.mathnet.ru/eng/vmp279 https://www.mathnet.ru/eng/vmp/v15/i4/p631
|
Statistics & downloads: |
Abstract page: | 290 | Full-text PDF : | 247 | References: | 1 |
|