Heidelberg 2022 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
T: Fachverband Teilchenphysik
T 69: DAQ and Trigger 3
T 69.7: Talk
Wednesday, March 23, 2022, 17:45–18:00, T-H28
Implementation of tracking algorithms for live reconstruction using AI processors — •Patrick Schwäbig1, Jochen Kaminski1, Michael Lupberger1, Klaus Desch1, and Stephen Neuendorffer2 — 1Physikalisches Institut, Universität Bonn, Deutschland — 2Xilinx Research Labs, San Jose, USA
For years, data rates generated by modern detectors and the corresponding readout electronics exceeded by far the limits of bandwidth and data storage space available in many experiments. Using fast triggers to discard uninteresting and irrelevant events is a solution used to this day. FPGAs, ASICs or even directly the readout chip are programmed or designed to apply a fixed set of rules based on low level parameters for an event pre-selection.
Up until the last few years, live track reconstruction for triggering was rarely possible due to a conflict between processing time and the required trigger latency. With the emergence of novel fast and highly parallelized processors, targeted mainly at AI inference, attempts to sufficiently accelerate tracking algorithms become viable. The Xilinx Versal AI Series Adaptive Compute Acceleration Platform (ACAP) is one such technology and combines traditional FPGA and CPU resources with dedicated AI cores and a network on chip for fast memory access.
In this talk AI and non-AI algorithms for track reconstruction and especially their implementation on the Xilinx VCK190/Versal VC1902 Evaluation Kit for a dark photon experiment at the ELSA accelerator in Bonn will be shown and the expected performance will be discussed.