Allen: A High-Level Trigger on GPUs for LHCb

R Aaij; J Albrecht; M Belous; P Billoir; T Boettcher; A Brea Rodríguez; D Vom Bruch; D H Cámpora Pérez; A Casais Vidal; D C Craik; P Fernandez Declara; L Funke; V V Gligorov; B Jashal; N Kazeev; D Martínez Santos; F Pisani; D Pliushchenko; S Popov; R Quagliani; M Rangel; F Reiss; C Sánchez Mayordomo; R Schwemmer; M Sokoloff; H Stevens; A Ustyuzhanin; X Vilasís Cardona; M Williams

doi:10.1007/s41781-020-00039-7

Allen: A High-Level Trigger on GPUs for LHCb

R Aaij, J Albrecht, M Belous, P Billoir, T Boettcher, A Brea Rodríguez, D Vom Bruch, D H Cámpora Pérez^*, A Casais Vidal, D C Craik, P Fernandez Declara, L Funke, V V Gligorov, B Jashal, N Kazeev, D Martínez Santos, F Pisani, D Pliushchenko, S Popov, R QuaglianiM Rangel, F Reiss, C Sánchez Mayordomo, R Schwemmer, M Sokoloff, H Stevens, A Ustyuzhanin, X Vilasís Cardona, M Williams

^*Corresponding author for this work

Research output: Contribution to journal › Article › Academic › peer-review

Abstract

We describe a fully GPU-based implementation of the first level trigger for the upgrade of the LHCb detector, due to start data taking in 2021. We demonstrate that our implementation, named Allen, can process the 40 Tbit/s data rate of the upgraded LHCb detector and perform a wide variety of pattern recognition tasks. These include finding the trajectories of charged particles, finding proton-proton collision points, identifying particles as hadrons or muons, and finding the displaced decay vertices of long-lived particles. We further demonstrate that Allen can be implemented in around 500 scientific or consumer GPU cards, that it is not I/O bound, and can be operated at the full LHC collision rate of 30 MHz. Allen is the first complete high-throughput GPU trigger proposed for a HEP experiment.

Original language	English
Article number	7
Journal	Computing and software for big science
Volume	4
Issue number	1
DOIs	https://doi.org/10.1007/s41781-020-00039-7
Publication status	Published - 2020

Keywords

physics.ins-det
hep-ex

Access to Document

10.1007/s41781-020-00039-7Licence: CC BY

Cite this

Aaij, R., Albrecht, J., Belous, M., Billoir, P., Boettcher, T., Brea Rodríguez, A., Vom Bruch, D., Cámpora Pérez, D. H., Casais Vidal, A., Craik, D. C., Fernandez Declara, P., Funke, L., Gligorov, V. V., Jashal, B., Kazeev, N., Martínez Santos, D., Pisani, F., Pliushchenko, D., Popov, S., ... Williams, M. (2020). Allen: A High-Level Trigger on GPUs for LHCb. Computing and software for big science, 4(1), Article 7. https://doi.org/10.1007/s41781-020-00039-7

@article{9893970589a543cbab5ba28d20d179ea,

title = "Allen: A High-Level Trigger on GPUs for LHCb",

abstract = "We describe a fully GPU-based implementation of the first level trigger for the upgrade of the LHCb detector, due to start data taking in 2021. We demonstrate that our implementation, named Allen, can process the 40 Tbit/s data rate of the upgraded LHCb detector and perform a wide variety of pattern recognition tasks. These include finding the trajectories of charged particles, finding proton-proton collision points, identifying particles as hadrons or muons, and finding the displaced decay vertices of long-lived particles. We further demonstrate that Allen can be implemented in around 500 scientific or consumer GPU cards, that it is not I/O bound, and can be operated at the full LHC collision rate of 30 MHz. Allen is the first complete high-throughput GPU trigger proposed for a HEP experiment.",

keywords = "physics.ins-det, hep-ex",

author = "R Aaij and J Albrecht and M Belous and P Billoir and T Boettcher and {Brea Rodr{\'i}guez}, A and {Vom Bruch}, D and {C{\'a}mpora P{\'e}rez}, {D H} and {Casais Vidal}, A and Craik, {D C} and {Fernandez Declara}, P and L Funke and Gligorov, {V V} and B Jashal and N Kazeev and {Mart{\'i}nez Santos}, D and F Pisani and D Pliushchenko and S Popov and R Quagliani and M Rangel and F Reiss and {S{\'a}nchez Mayordomo}, C and R Schwemmer and M Sokoloff and H Stevens and A Ustyuzhanin and {Vilas{\'i}s Cardona}, X and M Williams",

note = "12 pages, 12 figures, 2 tables",

year = "2020",

doi = "10.1007/s41781-020-00039-7",

language = "English",

volume = "4",

journal = "Computing and software for big science",

issn = "2510-2044",

publisher = "Springer Nature",

number = "1",

}

Aaij, R, Albrecht, J, Belous, M, Billoir, P, Boettcher, T, Brea Rodríguez, A, Vom Bruch, D, Cámpora Pérez, DH, Casais Vidal, A, Craik, DC, Fernandez Declara, P, Funke, L, Gligorov, VV, Jashal, B, Kazeev, N, Martínez Santos, D, Pisani, F, Pliushchenko, D, Popov, S, Quagliani, R, Rangel, M, Reiss, F, Sánchez Mayordomo, C, Schwemmer, R, Sokoloff, M, Stevens, H, Ustyuzhanin, A, Vilasís Cardona, X & Williams, M 2020, 'Allen: A High-Level Trigger on GPUs for LHCb', Computing and software for big science, vol. 4, no. 1, 7. https://doi.org/10.1007/s41781-020-00039-7

TY - JOUR

T1 - Allen

T2 - A High-Level Trigger on GPUs for LHCb

AU - Aaij, R

AU - Albrecht, J

AU - Belous, M

AU - Billoir, P

AU - Boettcher, T

AU - Brea Rodríguez, A

AU - Vom Bruch, D

AU - Cámpora Pérez, D H

AU - Casais Vidal, A

AU - Craik, D C

AU - Fernandez Declara, P

AU - Funke, L

AU - Gligorov, V V

AU - Jashal, B

AU - Kazeev, N

AU - Martínez Santos, D

AU - Pisani, F

AU - Pliushchenko, D

AU - Popov, S

AU - Quagliani, R

AU - Rangel, M

AU - Reiss, F

AU - Sánchez Mayordomo, C

AU - Schwemmer, R

AU - Sokoloff, M

AU - Stevens, H

AU - Ustyuzhanin, A

AU - Vilasís Cardona, X

AU - Williams, M

N1 - 12 pages, 12 figures, 2 tables

PY - 2020

Y1 - 2020

N2 - We describe a fully GPU-based implementation of the first level trigger for the upgrade of the LHCb detector, due to start data taking in 2021. We demonstrate that our implementation, named Allen, can process the 40 Tbit/s data rate of the upgraded LHCb detector and perform a wide variety of pattern recognition tasks. These include finding the trajectories of charged particles, finding proton-proton collision points, identifying particles as hadrons or muons, and finding the displaced decay vertices of long-lived particles. We further demonstrate that Allen can be implemented in around 500 scientific or consumer GPU cards, that it is not I/O bound, and can be operated at the full LHC collision rate of 30 MHz. Allen is the first complete high-throughput GPU trigger proposed for a HEP experiment.

AB - We describe a fully GPU-based implementation of the first level trigger for the upgrade of the LHCb detector, due to start data taking in 2021. We demonstrate that our implementation, named Allen, can process the 40 Tbit/s data rate of the upgraded LHCb detector and perform a wide variety of pattern recognition tasks. These include finding the trajectories of charged particles, finding proton-proton collision points, identifying particles as hadrons or muons, and finding the displaced decay vertices of long-lived particles. We further demonstrate that Allen can be implemented in around 500 scientific or consumer GPU cards, that it is not I/O bound, and can be operated at the full LHC collision rate of 30 MHz. Allen is the first complete high-throughput GPU trigger proposed for a HEP experiment.

KW - physics.ins-det

KW - hep-ex

U2 - 10.1007/s41781-020-00039-7

DO - 10.1007/s41781-020-00039-7

M3 - Article

C2 - 33385105

SN - 2510-2044

VL - 4

JO - Computing and software for big science

JF - Computing and software for big science

IS - 1

M1 - 7

ER -