Cargando…
Novel features and GPU performance analysis for EM particle transport in the Celeritas code
<!--HTML-->Celeritas is a new computational transport code designed for high-performance simulation of high-energy physics detectors. This work describes some of its current capabilities and the design choices that enable the rapid development of efficient on-device physics. The abstract...
Autor principal: | |
---|---|
Lenguaje: | eng |
Publicado: |
2021
|
Materias: | |
Acceso en línea: | http://cds.cern.ch/record/2767262 |
_version_ | 1780971285386362880 |
---|---|
author | Johnson, Seth |
author_facet | Johnson, Seth |
author_sort | Johnson, Seth |
collection | CERN |
description | <!--HTML-->Celeritas is a new computational transport code designed for high-performance
simulation of high-energy physics detectors. This work describes some of its
current capabilities and the design choices that enable the rapid development
of efficient on-device physics. The abstractions that underpin the code design
facilitate low-level performance tweaks that require no changes to the
higher-level physics code. We evaluate a set of independent changes that
together yield an almost 40\% speedup over the original GPU code for a net
performance increase of $220\times$ for a single GPU over a single CPU running
8.4M tracks on a small demonstration physics app. |
id | cern-2767262 |
institution | Organización Europea para la Investigación Nuclear |
language | eng |
publishDate | 2021 |
record_format | invenio |
spelling | cern-27672622022-11-02T22:25:37Zhttp://cds.cern.ch/record/2767262engJohnson, SethNovel features and GPU performance analysis for EM particle transport in the Celeritas code25th International Conference on Computing in High Energy & Nuclear PhysicsConferences<!--HTML-->Celeritas is a new computational transport code designed for high-performance simulation of high-energy physics detectors. This work describes some of its current capabilities and the design choices that enable the rapid development of efficient on-device physics. The abstractions that underpin the code design facilitate low-level performance tweaks that require no changes to the higher-level physics code. We evaluate a set of independent changes that together yield an almost 40\% speedup over the original GPU code for a net performance increase of $220\times$ for a single GPU over a single CPU running 8.4M tracks on a small demonstration physics app.oai:cds.cern.ch:27672622021 |
spellingShingle | Conferences Johnson, Seth Novel features and GPU performance analysis for EM particle transport in the Celeritas code |
title | Novel features and GPU performance analysis for EM particle transport in the Celeritas code |
title_full | Novel features and GPU performance analysis for EM particle transport in the Celeritas code |
title_fullStr | Novel features and GPU performance analysis for EM particle transport in the Celeritas code |
title_full_unstemmed | Novel features and GPU performance analysis for EM particle transport in the Celeritas code |
title_short | Novel features and GPU performance analysis for EM particle transport in the Celeritas code |
title_sort | novel features and gpu performance analysis for em particle transport in the celeritas code |
topic | Conferences |
url | http://cds.cern.ch/record/2767262 |
work_keys_str_mv | AT johnsonseth novelfeaturesandgpuperformanceanalysisforemparticletransportintheceleritascode AT johnsonseth 25thinternationalconferenceoncomputinginhighenergynuclearphysics |