Multiple Precision Floating-Point Arithmetic on SIMD Processors
Author:
- Joris Van Der HoevenJHJoris Van Der Hoeven
- 1 July 2017
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 10636889,p. 2-9
Abstract
Current general purpose libraries for multiple precision floating-point arithmetic such as MPFR suffer from a large performance penalty with respect to hard-wired instructions. The performance gap tends to become even larger with the advent of wider SIMD arithmetic in both CPUs and GPUs. In this paper, we present efficient algorithms for multiple precision floating- point arithmetic that are suitable for implementations on SIMD processors. A.C.M. subject classification: G.1.0 Computer-arithmetic A.M.S. subject classification: 65Y04, 65T50, 68W30.Keywords
All Related Versions
This publication has 12 references indexed in Scilit:
- Even faster integer multiplicationJournal of Complexity, 2016
- Optimizing Modular Multiplication for NVIDIA's Maxwell GPUs2016
- Modern Computer Arithmetic2010
- Handbook of Floating-Point Arithmetic2010
- Fast Quadruple Precision Arithmetic Library on Parallel Computer SR11000/J22008
- MPFRACM Transactions on Mathematical Software, 2007
- Algorithms for arbitrary precision floating point arithmetic2002
- Schnelle Multiplikation großer ZahlenComputing, 1971
- A floating-point technique for extending the available precisionNumerische Mathematik, 1971
- The Fast Fourier Transform in a Finite FieldMathematics of Computation, 1971