Speaker
Description
Future experiments at Jefferson Lab (JLab) and the Electron-Ion Collider (EIC) are designed to leverage the highest achievable luminosities to probe hadronic structure with unprecedented precision. These measurements require the efficient identification of rare physics events in the presence of substantial backgrounds. To address these challenges, scalable SI/ML algorithms capable of efficient detector-level reconstruction and event selection are actively under development to satisfy the throughput and latency requirements for streaming readout DAQ.
In this work, existing data from Jefferson Lab (Jlab) acquired with the Neutral Particle Spectrometer, a PWO based electromagnetic calorimeter were used. Transformer-based and graph neural network (GNN) models were trained for cluster reconstruction. These models have been integrated into the JANA2 reconstruction framework which enables GPU-accelerated inference. Building on this work, similar ML architectures are applied to EIC simulation data for detector-hit clustering and and their applicability is being evaluated to the more complex detector environment.
Future efforts will focus on rate capability test under realistic experimental conditions. At Jlab, a beam test will be performed using the small-scale NPS prototype electromagnetic calorimeter. JLab FADC250 and VXS Trigger Processor (VTP) modules will provide the bandwidth required for streaming readout. As part of Genesis efforts, software and hardware from industry are explored to optimize high-bandwidth data transfer and GPU resource utilization. These efforts will provide an estimate of achievable throughput and latency for the streaming-readout system while evaluating the applicability of the SI/ML-based models under realistic experimental conditions.