Portable acceleration of CMS computing workflows with coprocessors as a service
arXiv:2402.15366 · doi:10.1007/s41781-024-00124-1
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.
Replaced with the published version. Added the journal reference and the DOI. All the figures and tables can be found at http://cms-results.web.cern.ch/cms-results/public-results/publications/MLG-23-001 (CMS Public Pages)
References in corpus (57)
- XGBoost: A Scalable Tree Boosting System
- Observation of a new particle in the search for the Standard Model Higgs boson with the ATLAS detector at the LHC
- Observation of a new boson at a mass of 125 GeV with the CMS experiment at the LHC
- The anti-k_t jet clustering algorithm
- FastJet user manual
- Particle-flow reconstruction and global event description with the CMS detector
- The CMS trigger system
- Description and performance of track and primary-vertex reconstruction with the CMS tracker
- Observation of a new boson with mass near 125 GeV in pp collisions at sqrt(s) = 7 and 8 TeV
- Extraction and validation of a new set of CMS PYTHIA8 tunes from underlying-event measurements
- ParticleNet: Jet Tagging via Particle Clouds
- Deep Learning and its Application to LHC Physics
- Search for high-mass dilepton resonances using 139 fb of collision data collected at 13 TeV with the ATLAS detector
- Jet Substructure at the Large Hadron Collider: A Review of Recent Advances in Theory and Machine Learning
- Performance of the CMS Level-1 trigger in proton-proton collisions at 13 TeV
- Electron and photon reconstruction and identification with the CMS experiment at the CERN LHC
- Performance of missing transverse momentum reconstruction in proton-proton collisions at 13 TeV using the CMS detector
- Jet Flavour Classification Using DeepJet
- Search for resonant and nonresonant new phenomena in high-mass dilepton final states at 13 TeV
- Pileup mitigation at CMS in 13 TeV data
- Search for new phenomena in events with an energetic jet and missing transverse momentum in collisions at TeV with the ATLAS detector
- Search for new resonances in mass distributions of jet pairs using 139 fb of collisions at TeV with the ATLAS detector
- Search for supersymmetry in proton-proton collisions at 13 TeV in final states with jets and missing transverse momentum
- Identification of heavy, energetic, hadronically decaying particles using machine-learning techniques
- Search for narrow and broad dijet resonances in proton-proton collisions at 13 TeV and constraints on dark matter mediators and other new particles
- Operation of the ATLAS trigger system in Run 2
- Search for high mass dijet resonances with a new background prediction method in proton-proton collisions at 13 TeV
- Learning representations of irregular particle-detector geometry with distance-weighted graph networks
- Machine and Deep Learning Applications in Particle Physics
- JEDI-net: a jet identification algorithm based on interaction networks
- Search for new particles in events with energetic jets and large missing transverse momentum in proton-proton collisions at 13 TeV
- Search for supersymmetry in final states with two oppositely charged same-flavor leptons and missing transverse momentum in proton-proton collisions at 13 TeV
- Allen: A high level trigger on GPUs for LHCb
- Search for charginos and neutralinos in final states with two boosted hadronically decaying bosons and missing transverse momentum in collisions at TeV with the ATLAS detector
- MLPF: Efficient machine-learned particle-flow reconstruction using graph neural networks
- Interaction networks for the identification of boosted decays
- The Tracking Machine Learning challenge : Accuracy phase
- FPGA-accelerated machine learning inference as a service for particle physics computing
- Identification of hadronic tau lepton decays using a deep neural network
- Combined searches for the production of supersymmetric top quark partners in proton-proton collisions at 13 TeV
- Search for direct pair production of sleptons and charginos decaying to two leptons and neutralinos with mass splittings near the -boson mass in TeV collisions with the ATLAS detector
- Object condensation: one-stage grid-free multi-object reconstruction in physics detectors, graph and image data
- Search for higgsinos decaying to two Higgs bosons and missing transverse momentum in proton-proton collisions at 13 TeV
- Mini-AOD: A New Analysis Data Format for CMS
- HEPCloud, a New Paradigm for HEP Facilities: CMS Amazon Web Services Investigation
- Real-time data processing in the ALICE High Level Trigger at the LHC
- Reconstructing particles in jets using set transformer and hypergraph prediction networks
- Bringing heterogeneity to the CMS software framework
- GPU-accelerated machine learning inference as a service for computing in neutrino experiments
- Search for squarks and gluinos in final states with hadronically decaying -leptons, jets, and missing transverse momentum using collisions at = 13 TeV with the ATLAS detector
- Tracking Triggers for the HL-LHC
- Machine Learning for Particle Flow Reconstruction at CMS
- GPU coprocessors as a service for deep learning inference in high energy physics
- Search for top squarks in events with a Higgs or boson using 139 fb of collision data at TeV with the ATLAS detector
- Improved particle-flow event reconstruction with scalable neural networks for current and future particle detectors
- Real-time data processing with GPUs in high energy physics
- Accelerating Machine Learning Inference with GPUs in ProtoDUNE Data Processing