1.
|
Developments in Performance and Portability for MadGraph5_aMC@NLO
/ Valassi, Andrea (CERN) ; Childers, Taylor (Argonne (main)) ; Field, Laurence (CERN) ; Hageboeck, Stefan (CERN) ; Hopkins, Walter (Argonne (main)) ; Mattelaer, Olivier (Cathol. U. Louvain (main)) ; Nichols, Nathan (Argonne (main)) ; Roiser, Stefan (CERN) ; Smith, David (CERN)
Event generators simulate particle interactions using Monte Carlo techniques, providing the primary connection between experiment and theory in experimental high energy physics. These software packages, which are the first step in the simulation worflow of collider experiments, represent approximately 5 to 20% of the annual WLCG usage for the ATLAS and CMS experiments. [...]
arXiv:2210.11122.-
2022-10-20 - 6 p.
- Published in : PoS: ICHEP2022 (2022) , pp. 212
Fulltext: document - PDF; 2210.11122 - PDF;
In : 41st International Conference on High Energy Physics (ICHEP 2022), Bologna, Italy, 6 - 13 Jul 2022, pp.212
|
|
2.
|
Madgraph5_aMC@NLO on GPUs and vector CPUs Experience with the first alpha release
/ Hageboeck, Stephan (CERN) ; Childers, Taylor (Argonne (main)) ; Hopkins, Walter (Argonne (main)) ; Mattelaer, Olivier (Cathol. U. Louvain (main)) ; Nichols, Nathan (Argonne (main)) ; Roiser, Stefan (CERN) ; Teig, Jørgen (CERN) ; Valassi, Andrea (CERN) ; Vuosalo, Carl (U. Wisconsin, Madison (main)) ; Wettersten, Zenny (CERN)
Madgraph5_aMC@NLO is one of the most-frequently used Monte-Carlo event generators at the LHC, and an important consumer of compute resources. The software has been reengineered to maintain the overall look and feel of the user interface while speeding up event generation on CPUs and GPUs. [...]
arXiv:2312.02898.-
2024 - 8 p.
- Published in : EPJ Web Conf. 295 (2024) 11013
Fulltext: document - PDF; 2312.02898 - PDF;
In : 26th International Conference on Computing in High Energy & Nuclear Physics, Norfolk, Virginia, Us, 8 - 12 May 2023, pp.11013
|
|
3.
|
Porting CMS Heterogeneous Pixel Reconstruction to Kokkos
/ Kortelainen, Matti J. (Fermilab) ; Kwok, Martin (Fermilab) ; Childers, Taylor (Argonne (main)) ; Strelchenko, Alexei (Fermilab) ; Wang, Yunsong (LBL, Berkeley)
/CMS Collaboration
Programming for a diverse set of compute accelerators in addition to the CPU is a challenge. Maintaining separate source code for each architecture would require lots of effort, and development of new algorithms would be daunting if it had to be repeated many times. [...]
arXiv:2104.06573; FERMILAB-CONF-21-063-SCD.-
2021 - 10 p.
- Published in : EPJ Web Conf.: 251 (2021) , pp. 03034
Fulltext: fermilab-conf-21-063-scd - PDF; document - PDF; 2104.06573 - PDF; External link: Fermilab Library Server
In : 25th International Conference on Computing in High-Energy and Nuclear Physics (CHEP), Online, Online, 17 - 21 May 2021, pp.03034
|
|
4.
|
|
Porting CMS Heterogeneous Pixel Reconstruction to Kokkos
/ Kortelainen, Matti (speaker) (Fermi National Accelerator Lab. (US))
Programming for a diverse set of compute accelerators in addition to the CPU is a challenge. Maintaining separate source code for each architecture would require lots of effort, and development of new algorithms would be daunting if it had to be repeated many times. [...]
2021 - 741.
Conferences; 25th International Conference on Computing in High Energy & Nuclear Physics
External links: Talk details; Event details
In : 25th International Conference on Computing in High Energy & Nuclear Physics
|
|
5.
|
|
6.
|
|
Design and engineering of a simplified GPU workflow execution for the Madgraph5_aMC@NLO event generator
/ Valassi, Andrea (speaker) (CERN)
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. [...]
2021 - 1956.
Conferences; 25th International Conference on Computing in High Energy & Nuclear Physics
External links: Talk details; Event details
In : 25th International Conference on Computing in High Energy & Nuclear Physics
|
|
7.
|
|
|
Speeding up Madgraph5 aMC@NLO through CPU vectorization and GPU offloading: towards a first alpha release
/ Valassi, A. (CERN) ; Childers, T. (Argonne) ; Field, L. (CERN) ; Hageböck, S. (CERN) ; Hopkins, W. (Argonne) ; Mattelaer, O. (Louvain U., CP3) ; Nichols, N. (Argonne) ; Roiser, S. (CERN) ; Smith, D. (CERN) ; Teig, J. (CERN) et al.
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. [...]
arXiv:2303.18244.
-
8 p.
Fulltext
|
|
8.
|
|
GPU acceleration of scientific applications: An update
/ Messmer, Peter (speaker) (NVIDIA) ; Orlotti, Edmondo (speaker) (NVIDIA)
Just five years ago, NVIDIA introduced CUDA, the Compute Unified Device
Architecture, which significantly simplified the development of scientific
applications on Graphics Processing Units (GPUs).
Since these early days of CUDA, both GPU hardware and software have made
tremendous progress in terms of programmability and ease of use, allowing an
ever increasing set of applications to benefit from GPUs.
Especially the latest generation of NVIDIA GPUs, Kepler, introduced many
features geared towards HPC applications and to ensure high performance and
ease of use.
In this seminar, we will give an overview of the Kepler architecture,
present the new features of CUDA 5 and show how they can be used to
accelerate scientific applications.
In addition, we will present CARMA, the CUDA on ARM development board
hosting a NVIDIA Tegra CPU and CUDA GPU, and show how to use these boards to
explore an exciting high-performance, low power consumption node
architecture.
About the speakers
Edmondo Orlotti - HPC Development Manager
Edmondo Orlotti is responsible for the business development of NVIDIA HPC
Solutions in Southern Europe, for Education and Research. He joined NVIDIA
in 2005 as marketing manager for the same region, then moving in 2008 to the
professional solutions division. [...]
2012 - 4939.
Computing Seminar
External link: Event details
In : GPU acceleration of scientific applications: An update
|
|
9.
|
Design and engineering of a simplified workflow execution for the MG5aMC event generator on GPUs and vector CPUs
/ Valassi, Andrea (CERN) ; Roiser, Stefan (CERN) ; Mattelaer, Olivier (Louvain U.) ; Hageboeck, Stephan (CERN)
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. [...]
arXiv:2106.12631.-
2021 - 17 p.
- Published in : EPJ Web Conf. 251 (2021) 03045
Fulltext: 2106.12631 - PDF; document - PDF;
In : 25th International Conference on Computing in High-Energy and Nuclear Physics (CHEP), Online, Online, 17 - 21 May 2021, pp.03045
|
|
10.
|
Acceleration beyond lowest order event generation. An outlook on further parallelism within MadGraph5_aMC@NLO
/ Wettersten, Zenny (CERN ; Vienna, OAW) ; Mattelaer, Olivier (Louvain U., CP3) ; Roiser, Stefan (CERN) ; Schöfbeck, Robert (Vienna, OAW) ; Valassi, Andrea (CERN)
An important area of high energy physics studies at the Large Hadron Collider (LHC) currently concerns the need for more extensive and precise comparison data. Important tools in this realm are event reweighing and evaluation of more precise next-to-leading order (NLO) processes via Monte Carlo event generators, especially in the context of the upcoming High Luminosity LHC. [...]
arXiv:2312.07440.-
2024 - 8 p.
- Published in : EPJ Web Conf.: 295 (2024) , pp. 10001
Fulltext: 2312.07440 - PDF; document - PDF;
In : 26th International Conference on Computing in High Energy & Nuclear Physics, Norfolk, Virginia, Us, 8 - 12 May 2023, pp.10001
|
|