Cargando…
Validation for CMS ECAL - Reconstruction using GPU
The CMS experiment at CERN processes millions of signals every second during collisions in the large hadron collider (LHC). Each detec-tor in the experiment has a specific purpose to track certain particles or determine their energy and momentum. The electromagnetic calorime-ter (ECAL) is specialized...
Autor principal: | |
---|---|
Lenguaje: | eng |
Publicado: |
2019
|
Materias: | |
Acceso en línea: | http://cds.cern.ch/record/2689610 |
_version_ | 1780963736005115904 |
---|---|
author | Alnasrallah, Eissa A M A M |
author_facet | Alnasrallah, Eissa A M A M |
author_sort | Alnasrallah, Eissa A M A M |
collection | CERN |
description | The CMS experiment at CERN processes millions of signals every second during collisions in the large hadron collider (LHC). Each detec-tor in the experiment has a specific purpose to track certain particles or determine their energy and momentum. The electromagnetic calorime-ter (ECAL) is specialized in determining photons and electrons energies. The signal, once detected by ECAL crystals, goes through several process-ing programs to translate it from an electric signal to an energy cluster. The processing of all this data has high computational cost. The use of graphics processing units (GPUs) to process this data in a parallel manner largely reduces those computational costs. Currently, CMS team is working on transferring the CMS ECAL reconstruction code to GPU. In this project, I worked on developing a validation code for the ECAL Re-cHit producer code which converts amplitudes in uncalibrated RecHits to energies of RecHits. The validation code compares the GPU code results with the CPU’s. The current results are presented in this report along with the required future work for this process. |
id | cern-2689610 |
institution | Organización Europea para la Investigación Nuclear |
language | eng |
publishDate | 2019 |
record_format | invenio |
spelling | cern-26896102019-09-30T06:29:59Zhttp://cds.cern.ch/record/2689610engAlnasrallah, Eissa A M A MValidation for CMS ECAL - Reconstruction using GPUPhysics in GeneralThe CMS experiment at CERN processes millions of signals every second during collisions in the large hadron collider (LHC). Each detec-tor in the experiment has a specific purpose to track certain particles or determine their energy and momentum. The electromagnetic calorime-ter (ECAL) is specialized in determining photons and electrons energies. The signal, once detected by ECAL crystals, goes through several process-ing programs to translate it from an electric signal to an energy cluster. The processing of all this data has high computational cost. The use of graphics processing units (GPUs) to process this data in a parallel manner largely reduces those computational costs. Currently, CMS team is working on transferring the CMS ECAL reconstruction code to GPU. In this project, I worked on developing a validation code for the ECAL Re-cHit producer code which converts amplitudes in uncalibrated RecHits to energies of RecHits. The validation code compares the GPU code results with the CPU’s. The current results are presented in this report along with the required future work for this process.CERN-STUDENTS-Note-2019-198oai:cds.cern.ch:26896102019-09-16 |
spellingShingle | Physics in General Alnasrallah, Eissa A M A M Validation for CMS ECAL - Reconstruction using GPU |
title | Validation for CMS ECAL - Reconstruction using GPU |
title_full | Validation for CMS ECAL - Reconstruction using GPU |
title_fullStr | Validation for CMS ECAL - Reconstruction using GPU |
title_full_unstemmed | Validation for CMS ECAL - Reconstruction using GPU |
title_short | Validation for CMS ECAL - Reconstruction using GPU |
title_sort | validation for cms ecal - reconstruction using gpu |
topic | Physics in General |
url | http://cds.cern.ch/record/2689610 |
work_keys_str_mv | AT alnasrallaheissaamam validationforcmsecalreconstructionusinggpu |