LHCb GPU acceleration project

A. Badalov; D. Cámpora; N. Neufeld; X. Vilasís-Cardona

doi:10.1088/1748-0221/11/01/P01001

LHCb GPU acceleration project

A. Badalov^*
, D. Cámpora
, N. Neufeld
, X. Vilasís-Cardona

^*Corresponding author for this work

La Salle International School of Commerce and Digital Economy

Research output: Indexed journal article › Article › peer-review

1 Citation (Scopus)

Abstract

The LHCb detector is due to be upgraded for processing high-luminosity collisions, which will increase data bandwidth to the event filter farm from 100 GB/s to 4 TB/s, encouraging us to look for new ways of accelerating Online reconstruction. The Coprocessor Manager is a new framework for integrating LHCb's existing computation pipelines with massively parallel algorithms running on GPUs and other accelerators. This paper describes the system and analyzes its performance.

Original language	English
Article number	P01001
Journal	Journal of Instrumentation
Volume	11
Issue number	1
DOIs	https://doi.org/10.1088/1748-0221/11/01/P01001
Publication status	Published - 5 Jan 2016

Keywords

Data processing methods
Software architectures (event data models, frameworks and databases)

Access to Document

10.1088/1748-0221/11/01/P01001

Cite this

@article{0721eeff9de144f9b295a21e5c6a608b,

title = "LHCb GPU acceleration project",

abstract = "The LHCb detector is due to be upgraded for processing high-luminosity collisions, which will increase data bandwidth to the event filter farm from 100 GB/s to 4 TB/s, encouraging us to look for new ways of accelerating Online reconstruction. The Coprocessor Manager is a new framework for integrating LHCb's existing computation pipelines with massively parallel algorithms running on GPUs and other accelerators. This paper describes the system and analyzes its performance.",

keywords = "Data processing methods, Software architectures (event data models, frameworks and databases)",

author = "A. Badalov and D. C{\'a}mpora and N. Neufeld and X. Vilas{\'i}s-Cardona",

note = "Publisher Copyright: {\textcopyright} CERN 2016.",

year = "2016",

month = jan,

day = "5",

doi = "10.1088/1748-0221/11/01/P01001",

language = "English",

volume = "11",

journal = "Journal of Instrumentation",

issn = "1748-0221",

publisher = "Institute of Physics",

number = "1",

}

TY - JOUR

T1 - LHCb GPU acceleration project

AU - Badalov, A.

AU - Cámpora, D.

AU - Neufeld, N.

AU - Vilasís-Cardona, X.

PY - 2016/1/5

Y1 - 2016/1/5

N2 - The LHCb detector is due to be upgraded for processing high-luminosity collisions, which will increase data bandwidth to the event filter farm from 100 GB/s to 4 TB/s, encouraging us to look for new ways of accelerating Online reconstruction. The Coprocessor Manager is a new framework for integrating LHCb's existing computation pipelines with massively parallel algorithms running on GPUs and other accelerators. This paper describes the system and analyzes its performance.

AB - The LHCb detector is due to be upgraded for processing high-luminosity collisions, which will increase data bandwidth to the event filter farm from 100 GB/s to 4 TB/s, encouraging us to look for new ways of accelerating Online reconstruction. The Coprocessor Manager is a new framework for integrating LHCb's existing computation pipelines with massively parallel algorithms running on GPUs and other accelerators. This paper describes the system and analyzes its performance.

KW - Data processing methods

KW - Software architectures (event data models, frameworks and databases)

UR - https://www.scopus.com/pages/publications/84956986727

U2 - 10.1088/1748-0221/11/01/P01001

DO - 10.1088/1748-0221/11/01/P01001

M3 - Article

AN - SCOPUS:84956986727

SN - 1748-0221

VL - 11

JO - Journal of Instrumentation

JF - Journal of Instrumentation

IS - 1

M1 - P01001

ER -

LHCb GPU acceleration project

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this