English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The numerical study of relativistic magnetohydrodynamics (MHD) plays acrucial role in high-energy astrophysics, but unfortunately is computationallydemanding, given the complex physics involved (high Lorentz factor flows,extreme magnetization, curved spacetimes near compact objects) and the largevariety of spatial scales needed to resolve turbulent motions. A great benefitcomes from the porting of existing codes running on standard processors toGPU-based platforms. However, this usually requires a drastic rewriting of theoriginal code, the use of specific languages like CUDA, and a complex analysisof data management and optimization of parallel processes. Here we describe theporting of the ECHO code for special and general relativistic MHD toaccelerated devices, simply based on native Fortran language built-inconstructs, especially 'do concurrent' loops, few OpenACC directives, and thestraightforward data management provided by the Unified Memory option of NVIDIAcompilers.Thanks to these very minor modifications to the original code, thenew version of ECHO runs at least 16 times faster on GPU platforms compared toCPU-based ones. The chosen benchmark is the 3D propagation of a relativisticMHD Alfv\'en wave, for which strong and weak scaling tests performed on theLEONARDO pre-exascale supercomputer at CINECA are provided (using up to 256nodes corresponding to 1024 GPUs, and over 14 billion cells). Finally, anexample of high-resolution relativistic MHD Alfv\'enic turbulence simulation isshown, demonstrating the potential for astrophysical plasmas of the newGPU-based version of ECHO.

A GPU-Accelerated Modern Fortran Version of the ECHO Code for Relativistic Magnetohydrodynamics