Speaker
Description
ypzhang1991@ihep.ac.cn, IHEP, China
On behalf of the JUNO Collaboration
The Jiangmen Underground Neutrino Observatory (JUNO) is a large-scale, multi-physics neutrino experiment that utilizes tens of thousands of photomultiplier tubes (PMTs) to achieve unprecedented energy resolution. Signals from the PMTs are processed by front-end readout electronics and converted into digital ADC waveforms. JUNO is interested in an energy range spanning from tens of keV to tens of GeV. The high event rate and massive raw data generated by PMT waveforms necessitate an online event classification (OEC) system to identify events based on physical characteristics, compress data volume, and handle unusual data acquisition situations. The OEC system preserves reconstructed time/charge (T/Q) information for all events while selectively storing full waveforms or features for events of interest, which are crucial for subsequent offline precision reconstruction. The software implementation of the OEC system features a multithreaded Low-Level Event Classification (LEC) module and a single-threaded High-Level Event Classification (HEC) module. Furthermore, a middleware layer has been developed to facilitate the integration of offline reconstruction algorithms into the online environment. This presentation shows the implementation of the OEC system.