 |
Reproducible Triangular Solvers for High-Performance Computing
Abstract : On modern parallel architectures, floating-point computations may become non-deterministic and, therefore, non-reproducible mainly due to non-associativity of floating-point operations. We propose an algorithm to solve dense triangular systems by leveraging the standard parallel triangular solver and our, recently introduced, multi-level exact summation approach. Finally, we present implementations of the proposed fast repro-ducible triangular solver and results on recent NVIDIA GPUs.
Cited literature [16 references]
https://hal.archives-ouvertes.fr/hal-01116588
Contributor : Roman Iakymchuk <>
Submitted on : Saturday, February 14, 2015 - 5:30:54 PM
Last modification on : Friday, January 8, 2021 - 5:40:03 PM
Long-term archiving on: : Thursday, May 28, 2015 - 3:05:44 PM
Contributor : Roman Iakymchuk <>
Submitted on : Saturday, February 14, 2015 - 5:30:54 PM
Last modification on : Friday, January 8, 2021 - 5:40:03 PM
Long-term archiving on: : Thursday, May 28, 2015 - 3:05:44 PM
File
trsv.pdf
Files produced by the author(s)