Supported by:
Synopsys
Facebook
Facebook
Intel

Tuesday June 15th, 2021 from 10:45 to 12:00 (CEST)

Session 4. Custom precision floating point mathematical libraries

Chair: Nathalie Revol

1. Acceleration of LU decomposition supporting double-double, triple-double, and quadruple-double precision floating-point arithmetic with AVX2
Tomonori Kouya
[Paper]


2. A Study of the Effects and Benefits of Custom-Precision Mathematical Libraries for HPC Codes
Matei Istoan, Emeric Brun, David Defour, Pablo de Oliveira Castro, Davide Mancusi, Eric Petit, Alan Vaquet
[Paper]


3. Manifest for an approximation exchange format
Nicolas Brunie
[Paper]