Erlangen 2026 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
HK: Fachverband Physik der Hadronen und Kerne
HK 35: Instrumentation VI
HK 35.1: Group Report
Thursday, March 19, 2026, 13:45–14:15, PHIL A 301
First-level Event Selector of CBM — Jan de Cuveland1,2, Dirk Hutter1,2, and •Andreas Redelbach1,2 for the CBM collaboration — 1Frankfurt Institute for Advanced Studies — 2Goethe University Frankfurt, Germany
In the upcoming CBM experiment at GSI/FAIR very high interaction rates with multiple free-streaming triggerless detectors create huge amounts of data which must be processed in real-time. The First-level Event Selector (FLES) serves as the central event selection system of CBM. It functions as a high-performance computer cluster performing the online analysis of physics data, including full event reconstruction, at the incoming design data rate. Combining data from approximately 5000 input links to self-contained, overlapping processing intervals and distributing them to compute nodes form the basis for subsequent steps of online reconstruction and event selection. Timeslice intervals can be built efficiently over a high-throughput InfiniBand network and distributed to online computing resources for full online event reconstruction and analysis in a heterogeneous HPC cluster system. This also includes specialized algorithms for efficient processing of timeslice intervals in 4-D, and finally selecting the events relevant for storage. This presentation summarizes the status of the CBM FLES project. A particular focus will be on the underlying design combining maximum performance and flexibility with minimum memory consumption. Also recent developments will be shown that have been successfully tested at the CBM predecessor experiment mCBM.
This work is supported by BMBF (05P21RFFC1).
Keywords: CBM; FLES; DAQ; HLT; Online Computing