Optimization of finite-differencing kernels for numerical relativity applications

Roberto Alfieri, Sebastiano Bernuzzi, Albino Perego, David Radice

Research output: Contribution to journalArticlepeer-review

1 Scopus citations

Abstract

A simple optimization strategy for the computation of 3D finite-differencing kernels on many-cores architectures is proposed. The 3D finite-differencing computation is split direction-by-direction and exploits two level of parallelism: in-core vectorization and multi-threads shared-memory parallelization. The main application of this method is to accelerate the high-order stencil computations in numerical relativity codes. Our proposed method provides substantial speedup in computations involving tensor contractions and 3D stencil calculations on different processor microarchitectures, including Intel Knight Landing.

Original languageEnglish (US)
Article number15
JournalJournal of Low Power Electronics and Applications
Volume8
Issue number2
DOIs
StatePublished - Jun 2018

All Science Journal Classification (ASJC) codes

  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'Optimization of finite-differencing kernels for numerical relativity applications'. Together they form a unique fingerprint.

Cite this