% IMPORTANT: The following is UTF-8 encoded. This means that in the presence
% of non-ASCII characters, it will not work with BibTeX 0.99 or older.
% Instead, you should use an up-to-date BibTeX implementation like “bibtex8” or
% “biber”.
@ARTICLE{Hayrapetyan:604603,
author = {Hayrapetyan, Aram and others},
collaboration = {{CMS Collaboration}},
title = {{P}ortable acceleration of {CMS} computing workflows with
coprocessors as a service},
reportid = {PUBDB-2024-01181, arXiv:2402.15366. CMS-MLG-23-001.
CERN-EP-2023-303},
year = {2024},
note = {Submitted to Computing and Software for Big Science. All
figures and tables can be found at
http://cms-results.web.cern.ch/cms-results/public-results/publications/MLG-23-001
(CMS Public Pages)},
abstract = {Computing demands for large scientific experiments, such as
the CMS experiment at the CERN LHC, will increase
dramatically in the next decades. To complement the future
performance increases of software running on central
processing units (CPUs), explorations of coprocessor usage
in data processing hold great potential and interest.
Coprocessors are a class of computer processors that
supplement CPUs, often improving the execution of certain
functions due to architectural design choices. We explore
the approach of Services for Optimized Network Inference on
Coprocessors (SONIC) and study the deployment of this
as-a-service approach in large-scale data processing. In the
studies, we take a data processing workflow of the CMS
experiment and run the main workflow on CPUs, while
offloading several machine learning (ML) inference tasks
onto either remote or local coprocessors, specifically
graphics processing units (GPUs). With experiments performed
at Google Cloud, the Purdue Tier-2 computing center, and
combinations of the two, we demonstrate the acceleration of
these ML algorithms individually on coprocessors and the
corresponding throughput improvement for the entire
workflow. This approach can be easily generalized to
different types of coprocessors and deployed on local CPUs
without decreasing the throughput performance. We emphasize
that the SONIC approach enables high coprocessor usage and
enables the portability to run workflows on different types
of coprocessors.},
keywords = {CMS (INSPIRE) / performance (INSPIRE) / computer: network
(INSPIRE) / programming (INSPIRE) / machine learning
(INSPIRE) / Grid computing (INSPIRE) / microprocessor
(INSPIRE) / multiprocessor: graphics (INSPIRE) / cloud
(INSPIRE) / data management (INSPIRE)},
cin = {CMS},
cid = {I:(DE-H253)CMS-20120731},
pnm = {611 - Fundamental Particles and Forces (POF4-611) / DFG
project G:(GEPRIS)390833306 - EXC 2121: Quantum Universe
(390833306) / HIDSS-0002 - DASHH: Data Science in Hamburg -
Helmholtz Graduate School for the Structure of Matter
$(2019_IVF-HIDSS-0002)$},
pid = {G:(DE-HGF)POF4-611 / G:(GEPRIS)390833306 /
$G:(DE-HGF)2019_IVF-HIDSS-0002$},
experiment = {EXP:(DE-H253)LHC-Exp-CMS-20150101},
typ = {PUB:(DE-HGF)25},
eprint = {2402.15366},
howpublished = {arXiv:2402.15366},
archivePrefix = {arXiv},
SLACcitation = {$\%\%CITATION$ = $arXiv:2402.15366;\%\%$},
doi = {10.3204/PUBDB-2024-01181},
url = {https://bib-pubdb1.desy.de/record/604603},
}