Mixed-precision finite element kernels and assembly: Rounding error analysis and hardware acceleration

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Croci, M., Wells, G. N.
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866929546756882432
author Croci, M.
Wells, G. N.
author_facet Croci, M.
Wells, G. N.
contents In this paper we develop the first fine-grained rounding error analysis of finite element (FE) cell kernels and assembly. The theory includes mixed-precision implementations and accounts for hardware-acceleration via matrix multiplication units, thus providing theoretical guidance for designing reduced- and mixed-precision FE algorithms on CPUs and GPUs. Guided by this analysis, we introduce hardware-accelerated mixed-precision implementation strategies which are provably robust to low-precision computations. Indeed, these algorithms are accurate to the lower-precision unit roundoff with an error constant that is independent from: the conditioning of FE basis function evaluations, the ill-posedness of the cell, the polynomial degree, and the number of quadrature nodes. Consequently, we present the first AMX-accelerated FE kernel implementations on Intel Sapphire Rapids CPUs. Numerical experiments demonstrate that the proposed mixed- (single/half-) precision algorithms are up to 60 times faster than their double precision equivalent while being orders of magnitude more accurate than their fully half-precision counterparts.
format Preprint
id arxiv_https___arxiv_org_abs_2410_12614
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Mixed-precision finite element kernels and assembly: Rounding error analysis and hardware acceleration
Croci, M.
Wells, G. N.
Numerical Analysis
Hardware Architecture
Mathematical Software
In this paper we develop the first fine-grained rounding error analysis of finite element (FE) cell kernels and assembly. The theory includes mixed-precision implementations and accounts for hardware-acceleration via matrix multiplication units, thus providing theoretical guidance for designing reduced- and mixed-precision FE algorithms on CPUs and GPUs. Guided by this analysis, we introduce hardware-accelerated mixed-precision implementation strategies which are provably robust to low-precision computations. Indeed, these algorithms are accurate to the lower-precision unit roundoff with an error constant that is independent from: the conditioning of FE basis function evaluations, the ill-posedness of the cell, the polynomial degree, and the number of quadrature nodes. Consequently, we present the first AMX-accelerated FE kernel implementations on Intel Sapphire Rapids CPUs. Numerical experiments demonstrate that the proposed mixed- (single/half-) precision algorithms are up to 60 times faster than their double precision equivalent while being orders of magnitude more accurate than their fully half-precision counterparts.
title Mixed-precision finite element kernels and assembly: Rounding error analysis and hardware acceleration
topic Numerical Analysis
Hardware Architecture
Mathematical Software
url https://arxiv.org/abs/2410.12614