Accelerated Charged Particle Tracking with Graph Neural Networks on FPGAs
Aneesh M. Heintz, Vesal Razavimaleki, Javier Mauricio Duarte, G. Dezoort, I. Ojalvo, S. J. Thais, Markus Julian Atkinson, Mark S Neubauer, L. Gray, S. Jindariani, Nhan Viet Tran, Philip Coleman Harris, Dylan Sheldon Rankin, T. K. Aarrestad, Vladimir Lončar, M. Pierini, S. Summers, Jennifer Ngadiuba, Mia Liu, Edward Kreinar · arXiv (Cornell University) · 2020
We develop and study FPGA implementations of algorithms for charged particle tracking based on graph neural networks. The two complementary FPGA designs are based on OpenCL, a framework for writing programs that execute across heterogeneous platforms, and hls4ml, a high-level-synthesis-based compiler for neural network to firmware conversion. We evaluate and compare the resource usage, latency, and tracking performance of our implementations based on a benchmark dataset. We find a considerable speedup over CPU-based execution is possible, potentially enabling such algorithms to be used effectively in future computing workflows and the FPGA-based Level-1 trigger at the CERN Large Hadron Collider.