A New PIT Nb3Sn Process Toward Improved Cost-Performance for HEP High Field Magnets

Next generation High Energy Physics (HEP) applications consider high magnetic fields (>12 T) at liquid helium temperatures as well as low magnetic fields (/spl sim/2 T) and temperatures above 30 K. Multifilament Bi-2223 conductors have achieved the performance to be considered for these applications. In the low temperature, high magnetic field situation, Bi-2223 offers unique advantages over Nb/sub 3/Sn. Issues of Jc, field quality, magnet stability, strain tolerance, and manufacturability are discussed for accessing the merits of Bi-2223 as compared to Bi-2212 and A-15 materials for HEP.

High energy physics (HEP) experiments are facing unprecedented challenges to deal with huge volumes of data. Grid is the excellent technology for it. However, Incompatibility of a variety of grid middewares is an obstacle to exploit the resources on multiple grid infrastructures. This paper firstly introduces HEP experiments and HEP grid, and then reviews the solutions for grid interoperability and interoperations. Because current HEP applications are tightly with gLite middleware, this paper highlights a gateway approach to interoperate with gLite and described it in detail. Direct access to gLite storage system is very important, and a dedicated tool for HEP data sharing is then proposed and implemented. Finally, the implementation of the interoperability of GOS and gLite is introduced.

A high gain image intensifier system with fast shutter action for application in high energy physics

IntroductionReconstructing low-level particle tracks in neutrino physics can address some of the most fundamental questions about the universe. However, processing petabytes of raw data using deep learning techniques poses a challenging problem in the field of High Energy Physics (HEP). In the Exa.TrkX Project, an illustrative HEP application, preprocessed simulation data is fed into a state-of-art Graph Neural Network (GNN) model, accelerated by GPUs. However, limited GPU memory often leads to Out-of-Memory (OOM) exceptions during training, due to the large size of models and datasets. This problem is exacerbated when deploying models on High-Performance Computing (HPC) systems designed for large-scale applications.MethodsWe observe a high workload imbalance issue during GNN model training caused by the irregular sizes of input graph samples in HEP datasets, contributing to OOM exceptions. We aim to scale GNNs on HPC systems, by prioritizing workload balance in graph inputs while maintaining model accuracy. Our paper introduces diverse balancing strategies aimed at decreasing the maximum GPU memory footprint and avoiding the OOM exception, across various datasets.ResultsOur experiments showcase memory reduction of up to 32.14% compared to the baseline. We also demonstrate the proposed strategies can avoid OOM in application. Additionally, we create a distributed multi-GPU implementation using these samplers to demonstrate the scalability of these techniques on the HEP dataset.DiscussionBy assessing the performance of these strategies as data loading samplers across multiple datasets, we can gauge their effectiveness in both single-GPU and distributed environments. Our experiments, conducted on datasets of varying sizes and across multiple GPUs, broaden the applicability of our work to various GNN applications that handle input datasets with irregular graph sizes.

Niobium tin was one of the first superconductors to be made into relatively small high field magnets forty-six years ago. Conductors made from it are widely used for a range of different applications and development efforts are still being carried out to improve properties and reduce costs. The paper describes some of the work being carried out by the authors to make such improved conductors for fusion, high-energy physics and other applications. Three different types of niobium-tin conductors are described: 1. Low loss material with modest current density for fusion magnets. 2. High current density, higher loss material for high field high energy physics magnets. 3. Tubular type, low cost material with intermediate current density and small subelement size that could have application in a wide range of different magnets. Some of the advantages and drawbacks of each type of conductor are described together with illustrative properties.

SOFIE is a fast Machine Learning inference engine developed at CERN, capable of translating trained deep learning models—provided in ONNX, Keras, or PyTorch formats—into C++ code for efficient inference. The generated code has minimal dependencies, making it easily integrable into the data processing and analysis workflows of high-energy physics (HEP) experiments. This study presents a comprehensive benchmark analysis of SOFIE against leading machine learning frameworks for model evaluation, including PyTorch, TensorFlow XLA, and ONNX Runtime. We focus on evaluating their performance in HEP applications, particularly for typical models such as Graph Neural Networks for jet tagging, and Variational Autoencoders and Generative Adversarial Networks for fast simulation. Our assessment considers key factors such as computational speed, memory usage, scalability, and ease of integration with existing HEP software ecosystems. Through this comparative study, we aim to provide insights that help the HEP community select the most suitable framework for their specific needs.

Nb3Sn superconductor is of significant interest for applications in constructing high-field magnets beyond the limit of NbTi. However, its critical current density decreases rapidly at high magnetic fields (12 T) and the state-of-the-art level of Nb3Sn superconductors still cannot meet the requirements of the planned future accelerator magnets. The primary flux pinning centers for Nb3Sn wire mainly arise from the grain boundaries (GBs). In the present paper, we theoretically investigate, through time-dependent Ginzburg-Landau theory and with graphics-processing unit parallel technique, the vortex pinning and the critical current density in large-scale polycrystalline Nb3Sn superconductor by varying the pinning potential of GB and grain size at various magnetic fields. Unlike the conventional dot-like pinning systems, it is found that the critical current is not a monotonous function by suppressing the superconductivity of the GBs. The optimal pinning potential of GB for maximum critical current density strongly depends on magnetic fields. Furthermore, we find that the critical current density can be significantly enhanced by reducing grain size at low magnetic fields, while increase of critical current density cannot always be observed at high magnetic fields. Actually, critical current density even decreases by reducing grain size, which depends on the superconductivity of GBs. The findings in the paper provide theoretical foundations to achieve further improvement of Nb3Sn with optimizing the flux pinning.

Low-power amplifier-discriminators based on a so-called NINO architecture have been developed with high time resolution for the readout of radiation detectors. Two different circuits were integrated in the NINO13 chip, processed in IBM 130 nm CMOS technology. The LCO version (Low Capacitance and consumption Optimization) was designed for potential use as front-end electronics in the Gigatracker of the NA62 experiment at CERN. It was developed as pixel readout for solid-state pixel detectors to permit minimum ionizing particle detection with less than 180 ps rms resolution per pixel on the output pulse, for power consumption below 300 μW per pixel. The HCO version (High Capacitance Optimization) was designed with 4 mW power consumption per channel to provide timing resolution below 20 ps rms on the output pulse, for charges above 10 fC. Results presented show the potential of the LCO and HCO circuits for the precise timing readout of solid-state detectors, vacuum tubes or gas detectors, for applications in high energy physics, biotechnologies or medical imaging.

Quantum machine learning could possibly become a valuable alternative to classical machine learning for applications in high energy physics by offering computational speedups. In this study, we employ a support vector machine with a quantum kernel estimator (QSVM-Kernel method) to a recent LHC flagship physics analysis: $t\overline{t}H$ (Higgs boson production in association with a top quark pair). In our quantum simulation study using up to 20 qubits and up to $50\phantom{\rule{0.16em}{0ex}}000$ events, the QSVM-Kernel method performs as well as its classical counterparts in three different platforms from Google Tensorflow Quantum, IBM Quantum, and Amazon Braket. Additionally, using 15 qubits and 100 events, the application of the QSVM-Kernel method on the IBM superconducting quantum hardware approaches the performance of a noiseless quantum simulator. Our study confirms that the QSVM-Kernel method can use the large dimensionality of the quantum Hilbert space to replace the classical feature space in realistic physics data sets.

The time-projection chamber (TPC) is the main tracking detector in the MPD at NICA.Information on charge particle tracks is registered by the multi-wire proportional chamber with cathode pad readout.The main parameters and some details of the TPC design and readout electronics are presented.

In this paper, we show that affine extensions of non-crystallographic Coxeter groups can be derived via Coxeter-Dynkin diagram foldings and projections of affine extended versions of the root systems E8, D6, and A4. We show that the induced affine extensions of the non-crystallographic groups H4, H3, and H2 correspond to a distinguished subset of those considered in [P.-P. Dechant, C. Bœhm, and R. Twarock, J. Phys. A: Math. Theor. 45, 285202 (2012)]. This class of extensions was motivated by physical applications in icosahedral systems in biology (viruses), physics (quasicrystals), and chemistry (fullerenes). By connecting these here to extensions of E8, D6, and A4, we place them into the broader context of crystallographic lattices such as E8, suggesting their potential for applications in high energy physics, integrable systems, and modular form theory. By inverting the projection, we make the case for admitting different number fields in the Cartan matrix, which could open up enticing possibilities in hyperbolic geometry and rational conformal field theory.

Comparison of Bayesian and particle swarm algorithms for hyperparameter optimisation in machine learning applications in high energy physics

A hybrid photodetector (HPD) combined with a III-V photocathode is considered to be an ideal photodetector, because the quantum efficiency (QE) is high, the collection efficiency is 100%, and the excess noise factor is negligible. An HPD combined with a GaAs photocathode or a blue enhanced GaAs (BE-GaAs) photocathode has been developed. Presently, the QE is approximately 20% from 550 nm to 850 nm, which is much higher than that of a conventional multi-alkali (S-20) photocathode. The 100% collection efficiency calculated by the computer simulation is supported by the test result of the flat uniformity. Considering the application in high energy physics using a wavelength shifter, energy resolutions of both the BE-GaAs-HPD and the conventional S20-HPD were measured using a 650 nm LED light pulse. The BE-GaAs-HPD shows a resolution of 17%, as against 34% for the S20-HPD. The experimental results are consistent with the theoretical estimation. The QE of the HPD is especially important, because the electron multiplication process is noiseless.

Design study of a low-power, low-noise front-end for multianode silicon drift detectors

The aim of this paper is to fully characterize the new multi-anode photomultiplier tube R11265-103-M64, produced by Hamamatsu. Its high effective active area (77%), its pixel size, the low dark signal rate and the capability to detect single photon signals make this tube suitable for an application in high energy physics, such as for RICH detectors. Four tubes and two different bias voltage dividers have been tested. The results of a standard characterization of the gain and the anode uniformity, the dark signal rate, the cross-talk and the device behaviour as a function of temperature have been studied. The behaviour of the tube is studied in a longitudinal magnetic field up to 100 Gauss. Shields made of a high permeability material are also investigated. The deterioration of the device performance due to long time operation at intense light exposure is studied. A quantitative analysis of the variation of the gain and the dark signals rate due to the aging is described.