%0 Conference Paper
%A Bocci, Andrea
%A Das, Abhishek
%A Hatakeyama, Kenichi
%A Pantaleo, Felice
%A Lorkowski, Florian
%A Redjeb, Wahid
%A Samudio, Jonathan
%A Saunders, Mark
%T Heterogeneous reconstruction of hadronic particle flow clusters with the Alpaka Portability Library
%J The European physical journal / Web of Conferences
%V 337
%@ 2100-014X
%C Les Ulis
%I EDP Sciences
%M PUBDB-2026-00386
%P 01171 
%D 2025
%X In response to increasing data challenges, CMS has adopted the use of GPU offloading at the High-Level Trigger (HLT). However, GPU code is often hardware specific, and increases the maintenance burden on software development. The Alpaka (Abstraction Library for Parallel Kernel Acceleration) portability library offers a solution to this issue, and has been implemented into the CMS software (CMSSW) for use online at HLT. A portion of the final-state particle candidate reconstruction algorithm, Particle Flow, represented a target for increased performance through parallel operation. We discuss the port of hadronic Particle Flow clustering to Alpaka, and the validation of physics and performance at HLT for 2024 data taking.
%B 27th International Conference on Computing in High Energy and Nuclear Physics
%C 21 Oct 2024 - 25 Oct 2024, Krakow (Poland)
Y2 21 Oct 2024 - 25 Oct 2024
M2 Krakow, Poland
%F PUB:(DE-HGF)16 ; PUB:(DE-HGF)8
%9 Journal ArticleContribution to a conference proceedings
%R 10.1051/epjconf/202533701171
%U https://bib-pubdb1.desy.de/record/644527