Heidelberg 2022 – wissenschaftliches Programm

Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

T: Fachverband Teilchenphysik

T 25: Data Analysis, Information Technology and Artificial Intelligence

T 25.8: Vortrag

Montag, 21. März 2022, 18:10–18:25, T-H38

Information visualization platform for data quality monitoring of CMS tracker — •Abhit Patil — Ruprecht-Karls-Universität Heidelberg, Germany

The tracker of the CMS detector consists of silicon sensors arranged in concentric cylinders and endcap disks to track muons, which requires continuous monitoring during operation and certification of the recorded data for physics analysis. The process relies on shifters who assess the data quality by comparing data distributions with references. This challenging task requires examining possible types of failures with expert-based rule systems and manual profiling of a large number of histograms. To assess the quality of data volumes with finer granularity and to improve the quality of the data certification, this work proposes to augment the monitoring process with information visualization based methods, which aims to pre-process large amount of multidimensional data during the data taking period and provide a visual abstraction of the data quality and provide hints for potential anomalies. The visualisation methods are deployed on a platform built using Python-Django framework and Postgres database.