CERN Accelerating science

CMS Detector Performance Summaries

Darreres entrades:
2024-10-28
13:35
Jet tagging performance in Run 3 PbPb collisions at 5.36 TeV in CMS
The identification of jets initiated by heavy-flavoured particles is a crucial tool in the study of the interaction of these particles with the medium formed after a collision of heavy ions. The algorithms used in the first lead-lead (PbPb) LHC runs relied on the identification of secondary vertices, and the measurement of the displacement and other properties of the tracks in the jets. [...]
CMS-DP-2024-088; CERN-CMS-DP-2024-088.- Geneva : CERN, 2024 - 14 p. Fulltext: PDF;

Registre complet - Registres semblants
2024-10-24
16:59
Validation and monitoring of Geant4 simulation for CMS Phase2 detector configuration /CMS Collaboration
The CMS Simulation application performance for the CMS Phase-2 is presented, which includes results on CPU monitoring, results on physics validation for the combined test beam setup 2006 with barrel ecal and HCal calorimetry modules, and results on HGCal test beam 2018..
CMS-DP-2024-087; CERN-CMS-DP-2024-087.- Geneva : CERN, 2024 - 16 p. Fulltext: PDF;

Registre complet - Registres semblants
2024-10-22
09:44
Quantifying the computational speedup with madgraph4gpu for CMS worflow /CMS Collaboration
In this note, we present the first analysis of the computational speedup achieved using the GPU version of MadGraph, known as Madgraph4GPU(MG4GPU), in the CMS workflow. MadGraph is one of the most widely used event generators in CMS. [...]
CMS-DP-2024-086; CERN-CMS-DP-2024-086.- Geneva : CERN, 2024 - 23 p. Fulltext: PDF;

Registre complet - Registres semblants
2024-10-22
09:44
Update of the vertex reconstruction using track time from MTD /CMS Collaboration
The latest improvements in the vertex reconstruction using the timing information provided by the Mip Timing Detector (MTD) [1] are presented. An update of the vertex reconstruction algorithm with timing, labelled as ``4D'', is introduced and compared to the legacy algorithm, referred to as ``4DLegacy'', and to the 3D vertex algorithm, labelled as ``3Dt''. [...]
CMS-DP-2024-085; CERN-CMS-DP-2024-085.- Geneva : CERN, 2024 - 13 p. Fulltext: PDF;

Registre complet - Registres semblants
2024-10-21
10:52
A Multi-Objective Optimization Tool for HLT Pixel Track Reconstruction in CMS /CMS Collaboration
This note reports on the usage of an automatic parameter tuning algorithm for pixel track reconstruction in the context of HLT at LHC Run-3..
CMS-DP-2024-084; CERN-CMS-DP-2024-084.- Geneva : CERN, 2024 - 34 p. Fulltext: PDF;

Registre complet - Registres semblants
2024-10-21
10:52
Validation of the Alpaka-based heterogeneous reconstruction of pixel tracks and vertices /CMS Collaboration
This note reports the validation results of the reconstruction of HLT pixel tracks and vertices using either CUDA or Alpaka..
CMS-DP-2024-083; CERN-CMS-DP-2024-083.- Geneva : CERN, 2024 - 13 p. Fulltext: PDF;

Registre complet - Registres semblants
2024-10-21
10:52
2024 HLT timing and throughput results /CMS Collaboration
This note presents measurements of the computational performance of the HLT in 2024, in terms of timing (time per event), throughput (events processed per unit time), memory usage and power efficiency. Different configurations are compared, demonstrating the improvements brought by offloading part of the HLT reconstruction to GPUs..
CMS-DP-2024-082; CERN-CMS-DP-2024-082.- Geneva : CERN, 2024 - 38 p. Fulltext: PDF;

Registre complet - Registres semblants
2024-10-21
10:52
Towards a GPU compatible electron seeding algorithm /CMS Collaboration
This note presents first results on the development of a parallelizable algorithm for building tracking seeds in the reconstruction of electrons..
CMS-DP-2024-081; CERN-CMS-DP-2024-081.- Geneva : CERN, 2024 - 7 p. Fulltext: PDF;

Registre complet - Registres semblants
2024-10-14
10:52
FlashSim /CMS Collaboration
FlashSim is a Machine Learning based, end-to-end event simulation framework. It is being actively developed and optimized to meet the increasing demands of simulated samples, both now and in the near HL-LHC future. It is a fully agnostic, not analysis specific framework, trained on CMS, Geant4-based Full Simultation, going from (NANO)GEN to NANOAODSIM at O(100) Hz speed, with the aim to deliver a dataset in NANOAOD format at FastSim (or better) accuracy. [...]
CMS-DP-2024-080; CERN-CMS-DP-2024-080.- Geneva : CERN, 2024 - 35 p. Fulltext: PDF;

Registre complet - Registres semblants
2024-09-16
13:24
The role of the RPCs in the CMS Level-1 Trigger during Phase-2 /CMS Collaboration
This document describes the main features of the Endcap Muon Track Finder algorithm, which is the Level 1 trigger algorithm in the endcaps. In the endcaps of the CMS experiment there are three subsystems: Cathode Strip Chambers , Gaseous Electron Multipliers, and Resistive Plate Chambers (RPCs). [...]
CMS-DP-2024-079; CERN-CMS-DP-2024-079.- Geneva : CERN, 2024 - 11 p. Fulltext: PDF;

Registre complet - Registres semblants