Hierarchical Recursive Precision for Accelerating Symmetric Linear Solves on MXUs

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Carrica, Vicki, Alomairy, Rabab, Ringoot, Evelyne, Edelman, Alan
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913174575382528
author Carrica, Vicki
Alomairy, Rabab
Ringoot, Evelyne
Edelman, Alan
author_facet Carrica, Vicki
Alomairy, Rabab
Ringoot, Evelyne
Edelman, Alan
contents Symmetric positive-definite system solvers based on Cholesky factorization are fundamental to many scientific applications, such as climate modeling. We present a portable, nested recursive mixed-precision solver designed for Matrix Processing Units (MXUs), including NVIDIA Tensor Cores (H200) and AMD Matrix Cores (MI300X), that assigns low-precision FP16 arithmetic to large off-diagonal blocks, while preserving high precision on diagonal blocks to ensure numerical stability. The solver is implemented in Julia, providing a high-level, hardware-agnostic interface. We demonstrate up to a 5.07x speedup relative to the diagonal-precision vendor baseline, with 100x better accuracy than pure half precision on H200, providing higher accuracy than low-precision at higher speed than high-precision. Positive performance trends are also observed on MI300X, demonstrating broad applicability across GPUs.
format Preprint
id arxiv_https___arxiv_org_abs_2601_08082
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Hierarchical Recursive Precision for Accelerating Symmetric Linear Solves on MXUs
Carrica, Vicki
Alomairy, Rabab
Ringoot, Evelyne
Edelman, Alan
Distributed, Parallel, and Cluster Computing
Emerging Technologies
Mathematical Software
Performance
Symmetric positive-definite system solvers based on Cholesky factorization are fundamental to many scientific applications, such as climate modeling. We present a portable, nested recursive mixed-precision solver designed for Matrix Processing Units (MXUs), including NVIDIA Tensor Cores (H200) and AMD Matrix Cores (MI300X), that assigns low-precision FP16 arithmetic to large off-diagonal blocks, while preserving high precision on diagonal blocks to ensure numerical stability. The solver is implemented in Julia, providing a high-level, hardware-agnostic interface. We demonstrate up to a 5.07x speedup relative to the diagonal-precision vendor baseline, with 100x better accuracy than pure half precision on H200, providing higher accuracy than low-precision at higher speed than high-precision. Positive performance trends are also observed on MI300X, demonstrating broad applicability across GPUs.
title Hierarchical Recursive Precision for Accelerating Symmetric Linear Solves on MXUs
topic Distributed, Parallel, and Cluster Computing
Emerging Technologies
Mathematical Software
Performance
url https://arxiv.org/abs/2601.08082