Scholarlink

Abstract <p>NICA/MPD megascience project is the only world-class relativistic nuclear physics experiment in Russia, scheduled to start in 2025. MPDRoot is an offline software framework designed for simulation, reconstruction, and physical analysis of data obtained from the MPD experiment, projected to be on the 100&#xa0;PB scale. At the core of MPDRoot’s computational engine is the particle tracking in the TPC chamber of the MPD detector, having the principal influence on the quality and speed of data analysis. In this paper, we present first benchmarks of the MPD implementation of A common tracking software (ACTS) algorithmic suite against the default MPDRoot’s reconstruction algorithms. We show, that the implementations of the ACTS combinatorial Kalman filter tracking and adaptive multi vertex finder algorithms have comparable performance already in current alpha version, unoptimized for the specifics of the MPD detector. It is expected, that the future developments of ACTS suite implementation for the MPD detector will outperform and replace current TPC reconstruction algorithms.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Implementation of ACTS into MPDRoot

  • 期刊论文

  • 发表日期：

    2025年10月25日

查看全文
  • S. Hnatič,
  • J. Buša Jr.,
  • A. Bychkov,
  • A. Krylov,
  • V. Krylov,
  • A. Moshkin,
  • O. Rogachevsky

摘要

Abstract

NICA/MPD megascience project is the only world-class relativistic nuclear physics experiment in Russia, scheduled to start in 2025. MPDRoot is an offline software framework designed for simulation, reconstruction, and physical analysis of data obtained from the MPD experiment, projected to be on the 100 PB scale. At the core of MPDRoot’s computational engine is the particle tracking in the TPC chamber of the MPD detector, having the principal influence on the quality and speed of data analysis. In this paper, we present first benchmarks of the MPD implementation of A common tracking software (ACTS) algorithmic suite against the default MPDRoot’s reconstruction algorithms. We show, that the implementations of the ACTS combinatorial Kalman filter tracking and adaptive multi vertex finder algorithms have comparable performance already in current alpha version, unoptimized for the specifics of the MPD detector. It is expected, that the future developments of ACTS suite implementation for the MPD detector will outperform and replace current TPC reconstruction algorithms.