Design and engineering of a simplified workflow execution for the MG5aMC event generator on GPUs and vector CPUs
arXiv:2106.12631 · doi:10.1051/epjconf/202125103045
Abstract
Physics event generators are essential components of the data analysis software chain of high energy physics experiments, and important consumers of their CPU resources. Improving the software performance of these packages on modern hardware architectures, such as those deployed at HPC centers, is essential in view of the upcoming HL-LHC physics programme. In this paper, we describe an ongoing activity to reengineer the Madgraph5_aMC@NLO physics event generator, primarily to port it and allow its efficient execution on GPUs, but also to modernize it and optimize its performance on vector CPUs. We describe the motivation, engineering process and software architecture design of our developments, as well as the current challenges and future directions for this project. This paper is based on our submission to vCHEP2021 in March 2021,complemented with a few preliminary results that we presented during the conference. Further details and updated results will be given in later publications.
arXiv comment17 pages, 6 figures, submitted to vCHEP2021 proceedings in EPJ Web of Conferences; minor changes to address comments from the EPJWOC reviewer
References in corpus (22)
- The automated computation of tree-level and next-to-leading order differential cross sections, and their matching to parton shower simulations
- An Introduction to PYTHIA 8.2
- FeynRules 2.0 - A complete toolbox for tree-level phenomenology
- MadGraph 5 : Going Beyond
- Matching NLO QCD computations and parton shower simulations
- UFO - The Universal FeynRules Output
- MadEvent: Automatic Event Generation with MadGraph
- A standard format for Les Houches Event Files
- TikZ-Feynman: Feynman diagrams with TikZ
- Color-flow decomposition of QCD amplitudes
- ALOHA: Automatic Libraries Of Helicity Amplitudes for Feynman diagram computations
- Challenges in Monte Carlo event generator software for High-Luminosity LHC
- Alpaka - An Abstraction Library for Parallel Kernel Acceleration
- Monte Carlo integration on GPU
- VegasFlow: accelerating Monte Carlo simulation across multiple hardware platforms
- Fast calculation of HELAS amplitudes using graphics processing unit (GPU)
- PDFFlow: parton distribution functions on GPU
- Calculation of HELAS amplitudes for QCD processes using graphics processing unit (GPU)
- Fast computation of MadGraph amplitudes on graphics processing unit (GPU)
- MadFlow: automating Monte Carlo simulation on GPU for particle physics processes
- Towards the automation of Monte Carlo simulation on GPU for particle physics processes
- Using HEP experiment workflows for the benchmarking and accounting of WLCG computing resources
Cited by in corpus (10)
- Event Generators for High-Energy Physics Experiments
- The MadNIS Reloaded
- Efficient phase-space generation for hadron collider event simulation
- Developments in Performance and Portability for MadGraph5_aMC@NLO
- Potential of the Julia programming language for high energy physics computing
- Short reasons for long vectors in HPC CPUs: a study based on RISC-V
- A Portable Parton-Level Event Generator for the High-Luminosity LHC
- Accelerating Berends-Giele recursion for gluons in arbitrary dimensions over finite fields
- Challenges for Monte Carlo generators
- Generators and the (Accelerated) Future