Cargando…
Graphical processors for HEP trigger systems
General-purpose computing on GPUs is emerging as a new paradigm in several fields of science, although so far applications have been tailored to employ GPUs as accelerators in offline computations. With the steady decrease of GPU latencies and the increase in link and memory throughputs, time is rip...
Autores principales: | , , , , , , , , , , , , , , , , , , , , |
---|---|
Lenguaje: | eng |
Publicado: |
2017
|
Materias: | |
Acceso en línea: | https://dx.doi.org/10.1016/j.nima.2016.06.043 http://cds.cern.ch/record/2291595 |
_version_ | 1780956425147645952 |
---|---|
author | Ammendola, R Biagioni, A Chiozzi, S Cotta Ramusino, A Di Lorenzo, S Fantechi, R Fiorini, M Frezza, O Lamanna, G Lo Cicero, F Lonardo, A Martinelli, M Neri, I Paolucci, P S Pastorelli, E Piandani, R Pontisso, L Rossetti, D Simula, F Sozzi, M Vicini, P |
author_facet | Ammendola, R Biagioni, A Chiozzi, S Cotta Ramusino, A Di Lorenzo, S Fantechi, R Fiorini, M Frezza, O Lamanna, G Lo Cicero, F Lonardo, A Martinelli, M Neri, I Paolucci, P S Pastorelli, E Piandani, R Pontisso, L Rossetti, D Simula, F Sozzi, M Vicini, P |
author_sort | Ammendola, R |
collection | CERN |
description | General-purpose computing on GPUs is emerging as a new paradigm in several fields of science, although so far applications have been tailored to employ GPUs as accelerators in offline computations. With the steady decrease of GPU latencies and the increase in link and memory throughputs, time is ripe for real-time applications using GPUs in high-energy physics data acquisition and trigger systems. We will discuss the use of online parallel computing on GPUs for synchronous low level trigger systems, focusing on tests performed on the trigger of the CERN NA62 experiment. Latencies of all components need analysing, networking being the most critical. To keep it under control, we envisioned NaNet, an FPGA-based PCIe Network Interface Card (NIC) enabling GPUDirect connection. Moreover, we discuss how specific trigger algorithms can be parallelised and thus benefit from a GPU implementation, in terms of increased execution speed. Such improvements are particularly relevant for the foreseen LHC luminosity upgrade where highly selective algorithms will be crucial to maintain sustainable trigger rates with very high pileup. |
id | oai-inspirehep.net-1513619 |
institution | Organización Europea para la Investigación Nuclear |
language | eng |
publishDate | 2017 |
record_format | invenio |
spelling | oai-inspirehep.net-15136192019-09-30T06:29:59Zdoi:10.1016/j.nima.2016.06.043http://cds.cern.ch/record/2291595engAmmendola, RBiagioni, AChiozzi, SCotta Ramusino, ADi Lorenzo, SFantechi, RFiorini, MFrezza, OLamanna, GLo Cicero, FLonardo, AMartinelli, MNeri, IPaolucci, P SPastorelli, EPiandani, RPontisso, LRossetti, DSimula, FSozzi, MVicini, PGraphical processors for HEP trigger systemsDetectors and Experimental TechniquesComputing and ComputersGeneral-purpose computing on GPUs is emerging as a new paradigm in several fields of science, although so far applications have been tailored to employ GPUs as accelerators in offline computations. With the steady decrease of GPU latencies and the increase in link and memory throughputs, time is ripe for real-time applications using GPUs in high-energy physics data acquisition and trigger systems. We will discuss the use of online parallel computing on GPUs for synchronous low level trigger systems, focusing on tests performed on the trigger of the CERN NA62 experiment. Latencies of all components need analysing, networking being the most critical. To keep it under control, we envisioned NaNet, an FPGA-based PCIe Network Interface Card (NIC) enabling GPUDirect connection. Moreover, we discuss how specific trigger algorithms can be parallelised and thus benefit from a GPU implementation, in terms of increased execution speed. Such improvements are particularly relevant for the foreseen LHC luminosity upgrade where highly selective algorithms will be crucial to maintain sustainable trigger rates with very high pileup.oai:inspirehep.net:15136192017 |
spellingShingle | Detectors and Experimental Techniques Computing and Computers Ammendola, R Biagioni, A Chiozzi, S Cotta Ramusino, A Di Lorenzo, S Fantechi, R Fiorini, M Frezza, O Lamanna, G Lo Cicero, F Lonardo, A Martinelli, M Neri, I Paolucci, P S Pastorelli, E Piandani, R Pontisso, L Rossetti, D Simula, F Sozzi, M Vicini, P Graphical processors for HEP trigger systems |
title | Graphical processors for HEP trigger systems |
title_full | Graphical processors for HEP trigger systems |
title_fullStr | Graphical processors for HEP trigger systems |
title_full_unstemmed | Graphical processors for HEP trigger systems |
title_short | Graphical processors for HEP trigger systems |
title_sort | graphical processors for hep trigger systems |
topic | Detectors and Experimental Techniques Computing and Computers |
url | https://dx.doi.org/10.1016/j.nima.2016.06.043 http://cds.cern.ch/record/2291595 |
work_keys_str_mv | AT ammendolar graphicalprocessorsforheptriggersystems AT biagionia graphicalprocessorsforheptriggersystems AT chiozzis graphicalprocessorsforheptriggersystems AT cottaramusinoa graphicalprocessorsforheptriggersystems AT dilorenzos graphicalprocessorsforheptriggersystems AT fantechir graphicalprocessorsforheptriggersystems AT fiorinim graphicalprocessorsforheptriggersystems AT frezzao graphicalprocessorsforheptriggersystems AT lamannag graphicalprocessorsforheptriggersystems AT locicerof graphicalprocessorsforheptriggersystems AT lonardoa graphicalprocessorsforheptriggersystems AT martinellim graphicalprocessorsforheptriggersystems AT nerii graphicalprocessorsforheptriggersystems AT paoluccips graphicalprocessorsforheptriggersystems AT pastorellie graphicalprocessorsforheptriggersystems AT piandanir graphicalprocessorsforheptriggersystems AT pontissol graphicalprocessorsforheptriggersystems AT rossettid graphicalprocessorsforheptriggersystems AT simulaf graphicalprocessorsforheptriggersystems AT sozzim graphicalprocessorsforheptriggersystems AT vicinip graphicalprocessorsforheptriggersystems |