Göttingen 2025 – wissenschaftliches Programm

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

T: Fachverband Teilchenphysik

T 12: Data, AI, Computing, Electronics II (Data Management, Workflow)

T 12.1: Vortrag

Montag, 31. März 2025, 16:45–17:00, VG 2.102

Adaptation of the HammerCloud visualization to state-of-the-art tools - online and offline — •Lea Kuttler, Michael Böhler, and Markus Schumacher — Institute of Physics, Albert-Ludwigs-University Freiburg, Freiburg, Germany

HammerCloud (HC) is a framework for automated testing designed to monitor the resources of the Worldwide LHC Computing Grid (WLCG). It uses test jobs that mimic realistic physics analysis or production jobs to identify potential issues within the infrastructure of WLCG computing sites.

HC's web interface provides site administrators with detailed insights and visual summaries of test results. After almost two decades of successful operation, these visualizations, which previously relied on Google Image Charts, were upgraded to interactive Highcharts, offering a more dynamic, and user-friendly experience.

Complementing the detailed insights available on the HC website, the Tier-2-Report provides a broader overview by summarizing important test metrics over weekly or monthly periods to analyze overall performance trends. The generation of these Tier-2-Reports is fully automated, using a tool initially developed to present and discuss HC statistics for the GridKa cloud. A recent update has made this tool applicable for generating similar reports for any other cloud. Additionally, this update introduced a redesigned report layout and enhanced representations of test metrics.

This contribution presents the updated methods for visualizing and discussing test results within the HC framework.

Keywords: HammerCloud; WLCG; Distributed Computing