Speaker
Hyunjoon Kim
(SLAC National Accelerator Laboratory)
Description
In future high energy physics (HEP) experiments, proposed approaches to tracking the spatiotemporal information of charged particle trajectories on layers of detectors require FPGA clusters to process hit information and compute track parameters. For example, track triggering for the HL-LHC project requires processing of data volumes estimated up to 30 Tbits/s using 15K input combinatorics that define the track trigger decision within O(10μs) latency. This project aims to conduct a feasibility study on mapping the tracking problem to a custom-designed quadratic unconstrained binary optimization (QUBO) solver ASIC while also exploring track reconstruction time-to-solution, task energy/area efficiency and implementation scalability.
Primary author
Hyunjoon Kim
(SLAC National Accelerator Laboratory)
Co-authors
Angelo Dragone
(SLAC National Accelerator Laboratory)
Ariel Schwartzman
(SLAC National Accelerator Laboratory)
Prof.
Bongjin Kim
(University of California Santa Barbara)
Ms
Jooyoung Bae
(University of California Santa Barbara)
Lorenzo Rota
(SLAC National Accelerator Laboratory)
Dr
Rainer Bartoldus
(SLAC National Accelerator Laboratory)
