Speaker
Description
Significant computing resources are used for parton-level event generation for the Large Hadron Collider (LHC). The resource requirements of this part of the simulation toolchain are expected to grow further in the High-Luminosity (HL-LHC) era. At the same time, the rapid deployment of computing hardware different from the traditional CPU+RAM model in data centers around the world mandates a change in event generator design to provide sustainable simulations for the HL-LHC and future colliders.
We present the parton-level event generators Pepper and MadGraph4GPU, and discuss their performance and HPC scaling for providing expensive background samples at the LHC. We further showcase current developments, such as ML-optimised phase-space generation optimised by Normalizing Flow models, higher-order calculations, and physics-driven improvements of the numerical stability in infrared limits.
References
https://indico.cern.ch/event/1330797/contributions/5791236/
https://arxiv.org/abs/2406.07671
https://arxiv.org/abs/2311.06198
https://arxiv.org/abs/2503.21935
Significance
Beyond an introductory status report on portable event generation, I will present significant and novel computational and algorithmic advances in that context, namely the first application of Flow Matching to the problem of phase-space integration in particle physics providing significiant efficiency improvements, preliminary results on porting one-loop calculations to GPU (previous talks only covered leading-order calculations), and novel methods to increase very significantly the numerical stability in enhanced kinematic regions approaching the infrared singularities, which is a notorious problem already at NLO, but even more so at NNLO and beyond.
