CUMP library. CUMP  is a library for arbitrary precision arithmetic on CUDA , operating on floating point numbers. It is based on the GNU MP library (GMP) , and its functions have a GMP-like regular interface. CUMP is designed to be faster on NVIDIA GF100/GF110 GPUs than the GARPREC library . The performance advantage is achieved by using 64-bit fullwords as the basic arithmetic type, by using a register blocking technique, and by using ”little endian” word order format.
Keywords for this software
References in zbMATH (referenced in 1 article )
Showing result 1 of 1.
- Beliakov, Gleb; Matiyasevich, Yuri: A parallel algorithm for calculation of determinants and minors using arbitrary precision arithmetic (2016)