skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Speeding up particle track reconstruction using a parallel Kalman filter algorithm

Abstract

One of the most computationally difficult problems expected for the High-Luminosity Large Hadron Collider (HL-LHC) is determining the trajectory of charged particles during event reconstruction. Algorithms used at the LHC today rely on Kalman filtering, which builds physical trajectories incrementally while incorporating material effects and error estimation. Recognizing the need for faster computational throughput, we have adapted Kalman-filter-based methods for highly parallel, many-core SIMD architectures that are now prevalent in high-performance hardware. In this paper, we discuss the design and performance of the improved tracking algorithm, referred to as mkFit. A key piece of the algorithm is the Matriplex library, containing dedicated code to optimally vectorize operations on small matrices. The physics performance of the mkFit algorithm is comparable to the nominal CMS tracking algorithm when reconstructing tracks from simulated proton-proton collisions within the CMS detector. We study the scaling of the algorithm as a function of the parallel resources utilized and find large speedups both from vectorization and multi-threading. mkFit achieves a speedup of a factor of 6 compared to the nominal algorithm when run in a single-threaded application within the CMS software framework.

Authors:
; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;
Publication Date:
Research Org.:
Fermi National Accelerator Lab. (FNAL), Batavia, IL (United States)
Sponsoring Org.:
USDOE Office of Science (SC), High Energy Physics (HEP)
Contributing Org.:
CMS Collaboration
OSTI Identifier:
1638954
Report Number(s):
arXiv:2006.00071; FERMILAB-PUB-20-295-SCD
Journal ID: ISSN 1748-0221; oai:inspirehep.net:1798734
Grant/Contract Number:  
AC02-07CH11359
Resource Type:
Journal Article: Accepted Manuscript
Journal Name:
Journal of Instrumentation
Additional Journal Information:
Journal Volume: 15; Journal Issue: 09; Journal ID: ISSN 1748-0221
Publisher:
Institute of Physics (IOP)
Country of Publication:
United States
Language:
English
Subject:
46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND TECHNOLOGY; 72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS

Citation Formats

Lantz, S., McDermott, K., Reid, M., Riley, D., Wittich, P., Berkman, S., Cerati, G., Kortelainen, M., Hall, A. Reinsvold, Elmer, P., Wang, B., Giannini, L., Krutelyov, V., Masciovecchio, M., Tadel, M., Würthwein, F., Yagil, A., Gravelle, B., and Norris, B. Speeding up particle track reconstruction using a parallel Kalman filter algorithm. United States: N. p., 2020. Web. doi:10.1088/1748-0221/15/09/p09030.
Lantz, S., McDermott, K., Reid, M., Riley, D., Wittich, P., Berkman, S., Cerati, G., Kortelainen, M., Hall, A. Reinsvold, Elmer, P., Wang, B., Giannini, L., Krutelyov, V., Masciovecchio, M., Tadel, M., Würthwein, F., Yagil, A., Gravelle, B., & Norris, B. Speeding up particle track reconstruction using a parallel Kalman filter algorithm. United States. doi:10.1088/1748-0221/15/09/p09030.
Lantz, S., McDermott, K., Reid, M., Riley, D., Wittich, P., Berkman, S., Cerati, G., Kortelainen, M., Hall, A. Reinsvold, Elmer, P., Wang, B., Giannini, L., Krutelyov, V., Masciovecchio, M., Tadel, M., Würthwein, F., Yagil, A., Gravelle, B., and Norris, B. Tue . "Speeding up particle track reconstruction using a parallel Kalman filter algorithm". United States. doi:10.1088/1748-0221/15/09/p09030.
@article{osti_1638954,
title = {Speeding up particle track reconstruction using a parallel Kalman filter algorithm},
author = {Lantz, S. and McDermott, K. and Reid, M. and Riley, D. and Wittich, P. and Berkman, S. and Cerati, G. and Kortelainen, M. and Hall, A. Reinsvold and Elmer, P. and Wang, B. and Giannini, L. and Krutelyov, V. and Masciovecchio, M. and Tadel, M. and Würthwein, F. and Yagil, A. and Gravelle, B. and Norris, B.},
abstractNote = {One of the most computationally difficult problems expected for the High-Luminosity Large Hadron Collider (HL-LHC) is determining the trajectory of charged particles during event reconstruction. Algorithms used at the LHC today rely on Kalman filtering, which builds physical trajectories incrementally while incorporating material effects and error estimation. Recognizing the need for faster computational throughput, we have adapted Kalman-filter-based methods for highly parallel, many-core SIMD architectures that are now prevalent in high-performance hardware. In this paper, we discuss the design and performance of the improved tracking algorithm, referred to as mkFit. A key piece of the algorithm is the Matriplex library, containing dedicated code to optimally vectorize operations on small matrices. The physics performance of the mkFit algorithm is comparable to the nominal CMS tracking algorithm when reconstructing tracks from simulated proton-proton collisions within the CMS detector. We study the scaling of the algorithm as a function of the parallel resources utilized and find large speedups both from vectorization and multi-threading. mkFit achieves a speedup of a factor of 6 compared to the nominal algorithm when run in a single-threaded application within the CMS software framework.},
doi = {10.1088/1748-0221/15/09/p09030},
journal = {Journal of Instrumentation},
issn = {1748-0221},
number = 09,
volume = 15,
place = {United States},
year = {2020},
month = {9}
}

Journal Article:
Free Publicly Available Full Text
This content will become publicly available on September 22, 2021
Publisher's Version of Record

Save / Share:

Works referenced in this record:

LHC Machine
journal, August 2008


The CMS trigger system
journal, January 2017


Vertexing and Tracking Algoritms at High Pile-Up
conference, May 2015

  • Cerati, Giuseppe
  • Proceedings of The 23rd International Workshop on Vertex Detectors — PoS(Vertex2014)
  • DOI: 10.22323/1.227.0037

A Roadmap for HEP Software and Computing R&D for the 2020s
journal, March 2019

  • Albrecht, Johannes; Alves, Antonio Augusto; Amadio, Guilherme
  • Computing and Software for Big Science, Vol. 3, Issue 1
  • DOI: 10.1007/s41781-018-0018-8

Pileup per particle identification
journal, October 2014

  • Bertolini, Daniele; Harris, Philip; Low, Matthew
  • Journal of High Energy Physics, Vol. 2014, Issue 10
  • DOI: 10.1007/JHEP10(2014)059

Identification of heavy-flavour jets with the CMS detector in pp collisions at 13 TeV
journal, May 2018


Performance of reconstruction and identification of τ leptons decaying to hadrons and v τ in pp collisions at √ s =13 TeV
journal, October 2018


CMS tracking performance results from early LHC operation
journal, November 2010


Observation of the diphoton decay of the Higgs boson and measurement of its properties
journal, October 2014


Application of Kalman filtering to track and vertex fitting
journal, December 1987

  • Frühwirth, R.
  • Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, Vol. 262, Issue 2-3
  • DOI: 10.1016/0168-9002(87)90887-4

A New Approach to Linear Filtering and Prediction Problems
journal, March 1960

  • Kalman, R. E.
  • Journal of Basic Engineering, Vol. 82, Issue 1
  • DOI: 10.1115/1.3662552

First evaluation of the CPU, GPGPU and MIC architectures for real time particle tracking based on Hough transform at the LHC
journal, April 2014


Description and performance of track and primary-vertex reconstruction with the CMS tracker
journal, October 2014


Performance of the ATLAS track reconstruction algorithms in dense environments in LHC Run 2
journal, October 2017


Designing vector-friendly compact BLAS and LAPACK kernels
conference, January 2017

  • Kim, Kyungjoo; Costa, Timothy B.; Deveci, Mehmet
  • Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis on - SC '17
  • DOI: 10.1145/3126908.3126941

Review of Particle Physics
journal, August 2018


Precision measurement of the structure of the CMS inner tracking system using nuclear interactions
journal, October 2018


The Tau Parallel Performance System
journal, May 2006

  • Shende, Sameer S.; Malony, Allen D.
  • The International Journal of High Performance Computing Applications, Vol. 20, Issue 2
  • DOI: 10.1177/1094342006064482

Validity of the single processor approach to achieving large scale computing capabilities
conference, January 1967

  • Amdahl, Gene M.
  • Proceedings of the April 18-20, 1967, spring joint computer conference on - AFIPS '67 (Spring)
  • DOI: 10.1145/1465482.1465560