% 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},
}