Speeding up Madgraph5 aMC@NLO through CPU vectorization and GPU offloading: towards a first alpha release

TL;DR

This paper discusses efforts to accelerate the Madgraph5 aMC@NLO event generator by implementing CPU vectorization and GPU offloading for matrix element calculations, aiming for a first alpha release to improve LHC physics simulations.

Contribution

It introduces the reengineering of Madgraph5 aMC@NLO with vectorized C++, CUDA, and SYCL implementations for matrix element calculations, enhancing computational efficiency.

Findings

Successful implementation of vectorized C++ code for ME calculations

Development of CUDA and SYCL versions for GPU acceleration

Progress towards an alpha release supporting QCD LO processes

Abstract

The matrix element (ME) calculation in any Monte Carlo physics event generator is an ideal fit for implementing data parallelism with lockstep processing on GPUs and vector CPUs. For complex physics processes where the ME calculation is the computational bottleneck of event generation workflows, this can lead to large overall speedups by efficiently exploiting these hardware architectures, which are now largely underutilized in HEP. In this paper, we present the status of our work on the reengineering of the Madgraph5_aMC@NLO event generator at the time of the ACAT2022 conference. The progress achieved since our previous publication in the ICHEP2022 proceedings is discussed, for our implementations of the ME calculations in vectorized C++, in CUDA and in the SYCL framework, as well as in their integration into the existing MadEvent framework. The outlook towards a first alpha release of the software supporting QCD LO processes usable by the LHC experiments is also discussed.