Nicholas Manganelli

With machine learning applications now spanning a variety of computational tasks, multi-user shared computing facilities are devoting a rapidly increasing proportion of their resources to such algorithms. Graph neural networks (GNNs), for example, have provided astounding improvements in extracting complex signatures from data and are now widely used in a variety of applications, such as particle jet classification in high energy physics (HEP). However, GNNs also come with an enormous computational penalty that requires the use of GPUs to maintain reasonable throughput. At shared computing facilities, such as those used by physicists at Fermi National Accelerator Laboratory (Fermilab), methodical resource allocation and high throughput at the many-user scale are key to ensuring that resources are being used as efficiently as possible. These facilities, however, primarily provide CPU-only nodes, which proves detrimental to time-to-insight and computational throughput for workflows that include machine learning inference. In this work, we describe how a shared computing facility can use the NVIDIA Triton Inference Server to optimize its resource allocation and computing structure, recovering high throughput while scaling out to multiple users by massively parallelizing their machine learning inference. To demonstrate the effectiveness of this system in a realistic multi-user environment, we use the Fermilab Elastic Analysis Facility augmented with the Triton Inference Server to provide scalable and high throughput access to a HEP-specific GNN and report on the outcome.

The second workshop on the HEP Analysis Ecosystem took place 23-25 May 2022 at IJCLab in Orsay, to look at progress and continuing challenges in scaling up HEP analysis to meet the needs of HL-LHC and DUNE, as well as the very pressing needs of LHC Run 3 analysis. The workshop was themed around six particular topics, which were felt to capture key questions, opportunities and challenges. Each topic arranged a plenary session introduction, often with speakers summarising the state-of-the art and the next steps for analysis. This was then followed by parallel sessions, which were much more discussion focused, and where attendees could grapple with the challenges and propose solutions that could be tried. Where there was significant overlap between topics, a joint discussion between them was arranged. In the weeks following the workshop the session conveners wrote this document, which is a summary of the main discussions, the key points raised and the conclusions and outcomes. The document was circulated amongst the participants for comments before being finalised here.

The standard model production of four top quarks is predicted to have a cross section of the order of 12fb. The CMS Collaboration presents new results on this rare production mechanism for Run 2 data collected in 2016 through 2018 at 13 TeV, considering event signatures containing zero to four electrons or muons. This is the first time the all-hadronic channel is investigated in the study of four top quarks, made possible through novel machine learning based data-driven background estimation techniques.

The Large Hadron Collider (LHC) will be upgraded in several phases to significantly expand its physics program, and these upgrades present major challenges to the operations of the CMS Cathode Strip Chamber muon system. After the current long shutdown during 2018–2021 (LS2), the accelerator luminosity will have increased, exceeding the design value of 1034 cm−2s−1 and allowing the CMS experiment to collect approximately 100 fb−1/year. A subsequent upgrade in 2023–2024 will increase the luminosity up to 5 × 1034 cm−2s−1. The CMS muon system must be able to sustain a physics program after the LS2 shutdown that maintains and enhances sensitivity to electroweak scale physics and for TeV scale searches similar to what was achieved up to now. For the Cathode Strip Chamber (CSC) muon detectors, the electronics will be upgraded to handle the expected higher data rates. The design of the upgraded CSC electronics is discussed in this report. In addition, accelerated irradiation tests are being performed to study the behaviour of the CSC electronics under conditions that are nearly an order of magnitude beyond the original design values. Studies have also been performed of chamber gas mixtures to reduce greenhouse-gas impacts. The status of this irradiation campaign and results are presented.