Product Design, Manufacturing & Innovation Resources

Dernières publications et brevets sur le calcul à haute performance (HPC)

Calcul à haute performance

Astuce : suite à la sélection ci-dessous, vous pouvez rechercher et filtrer nos 2 bases de données complètes :

> outil de recherche de publications gratuit < par auteur, sujet, mots-clés, date ou revue.

> outil gratuit de recherche de brevets < pour les brevets en anglais de l'Office européen des brevets.

Calcul haute performance
Le calcul haute performance permet de faire progresser la conception et l'optimisation des systèmes informatiques complexes pour un traitement à très grande vitesse dans diverses applications.

High-performance computing focuses on the design, implementation, and optimization of computing systems capable of executing large-scale, complex computations at extreme speeds. It involves architectures such as clusters, supercomputers, and heterogeneous systems integrating CPUs and GPUs, alongside programming models like MPI and OpenMP to enable parallelism across distributed and shared memory environments. Key challenges include efficient resource management, load balancing, fault tolerance, and minimizing communication overhead within interconnect networks. This domain supports scientific simulations, data-intensive analytics, and real-time processing demands across disciplines. The following compilation presents recent publications and patents detailing advancements in hardware architectures, logiciel frameworks, algorithms, and performance optimization techniques in HPC.

Il s'agit de notre dernière sélection de publications et de brevets mondiaux en anglais sur le calcul à haute performance (HPC), parmi de nombreuses revues scientifiques en ligne, classées et axées sur le calcul à haute performance, HPC, calcul parallèle, calcul distribué, supercalculateur, calcul en grappe, accélération gpu, MPI, openMP, calcul à haut débit, équilibrage de charge, réseau d'interconnexion, planification des tâches, processus vectoriel, hiérarchie de la mémoire, cohérence du cache, tolérance aux fautes, parallélisme des données, parallélisme des tâches, mémoire distribuée, calcul exascale, équilibrage de charge, réseau d'interconnexion, planification des tâches, calcul hétérogène, processus vectoriel, hiérarchie de la mémoire, cohérence du cache, tolérance aux pannes, parallélisme des données, parallélisme des tâches, mémoire distribuée, calcul exascale, calcul petascale, optimisation des performances, système de fichiers parallèles, topologie du réseau, programmation d'accélérateurs et calcul à haute performance.

Method for calculating a multi-dimensional image-domain system matrix for mpi measurements using a multi-dimensional sinogram-based system matrix

Patent published on the 2026-05-21 in US under Ref US20260140207 by BRUKER BIOSPIN GMBH & CO KG [DE] (Ilbey Serhat [de], Franke Jochen [de])

Abstract: A magnetic particle imaging (MPI) measurement method includes generating a multi-dimensional image-domain system matrix A using an MPI-sequence with a field free line, using a multi-dimensional sinogram-based system matrixASinogram(t,r,z)∈ℝ[T,Nr,Nz],comprising the following steps: for each time point tc, c∈[1, T]: get a time slice matrix Ac=ASinogram(t=tc, r, z)∈[Nr,Nz],replicate the time slice matrix Ac Nr−1 times and copying these replications along a newly added 2nd offset-dimension[...]


Our summary: This method calculates a multi-dimensional image-domain system matrix for MPI measurements. It involves generating a time volume matrix from a sinogram-based system matrix and rotating it based on a field free line. Finally, a Fourier transformation is performed to obtain the constructed system matrix.

multi-dimensional image-domain, system matrix, magnetic particle imaging, Fourier transformation

Patent

Ultra fast pulser for low energy electron beams

Patent published on the 2026-05-21 in US under Ref US20260142118 by EUCLID TECHLABS LLC [US] (Jing Chunguang [us], Montgomery Eric John [us], Zhao Yubin [us], Wyderski Bartosz [us], Miller Sean [us], Rush Wade [us], Kostin Roman [us], Kanareykin Alexei [us])

Abstract: An electron beam pulser suitable for SEM systems directs ≤30 keV electrons through a stripline kicker. RF waves or voltage pulses deflect the beam to periodically impinge on an aperture, thereby creating electron pulses from 100 fs to 100 ns in width. In RF embodiments, the RF frequency is tuned to provide 180-degree phase slippage between the RF and the electron beam, thereby generating pulse repetition rates up to 20 GHz without divergence. The RF can be reverse propagated through the kicker[...]


Our summary: This device generates ultra-fast electron pulses for SEM systems using a stripline kicker. It achieves pulse widths from 100 fs to 100 ns with repetition rates up to 20 GHz. The system can be synchronized for advanced pump-probe experiments.

electron beam, pulser, RF waves, pulse width

Patent

System and method for adaptive delivery of chemicals to electric submersible pumping systems

Patent published on the 2026-05-21 in WO under Ref WO2026106744 by BAKER HUGHES OILFIELD OPERATIONS LLC [US] (Gaddy Dean [us], Adesanwo Moradeyo [us])

Abstract: Systems and methods are disclosed for carrying out a chemical treatment on a well located at a well site, wherein an electric submersible pumping system is deployed in the well to recover hydrocarbons and associated fluids. The method includes the steps of connecting a chemical treatment system to the well that includes an injection pump and an injection pump controller, initiating the chemical treatment by setting the injection pump at an initial flow rate, calculating a virtual flow rate for t[...]


Our summary: The system enables chemical treatment at a well site using an electric submersible pumping system. It connects an injection pump and controller to manage chemical delivery. The method adjusts the pump output based on a calculated virtual flow rate for optimal treatment.

adaptive delivery, chemical treatment, electric submersible pumping, flow rate adjustment

Patent

Discharging of a damping capacitor

Patent published on the 2026-05-21 in WO under Ref WO2026104097 by AMS OSRAM AG [AT] (Fiocchi Carlo [it], Schallmoser Oskar [de], Geiger Robert [de])

Abstract: The present disclosure relates to a circuit including: a supply terminal configured to receive a supply voltage; a driver circuit including an input terminal coupled with the supply terminal to receive the supply voltage as input voltage for powering an operation of the driver circuit; a damping capacitor coupled in parallel with the input terminal and ground such that the damping capacitor is charged by the supply voltage; wherein the driver circuit further comprises a discharge current generat[...]


Our summary: The circuit includes a supply terminal and a driver circuit powered by the supply voltage. A damping capacitor is charged in parallel with the input terminal and discharges when an overvoltage criterion is met. The discharge current generator controls the discharge current s behavior inversely to the input voltage.

damping capacitor, discharge current, overvoltage criterion, driver circuit

Patent

Crane assembly for use in vehicle glazing replacement

Patent published on the 2026-05-21 in WO under Ref WO2026104392 by BELRON INT LTD [GB] (Berry Luke [gb], Quayum Sarwar [gb])

Abstract: The present specification relates to a crane arrangement that can be used in the fitting of vehicle windows, particularly for vehicle windscreens. Described is a crane assembly for lifting a vehicle glazing panel relative to a vehicle during a vehicle glazing panel installation or removal process. The crane assembly comprises a clamp arrangement for clamping onto a vehicle wheel; a mast extending from the clamp arrangement; and a boom extending from the mast for supporting a vehicle glazing pane[...]


Our summary: The crane assembly is designed for lifting vehicle glazing panels during installation or removal. It features a clamp that attaches to a vehicle wheel, a mast, and a boom. This arrangement facilitates efficient and safe vehicle windscreen fitting.

Crane assembly, vehicle glazing, installation, clamp arrangement

Patent

System and method for synchronous aggregation and amalgamation of occupational specifications by intelligent neural networking

Patent published on the 2026-05-21 in US under Ref US20260141350 by THOMPSON II DARRELL [US] (Thompson Ii Darrell [us])

Abstract: [0000] The invention discloses systems and methods for synchronous aggregation and amalgamation of occupational specifications by intelligent neural gathering of US Department of Labor Standard Occupational Classification, and Occupational Information Network standardized data descriptions. The system constructs standardized data configurations of job canon via program-global area data files and cumulative career historical data using system-global area data files, with the purpose of producing [...]


Our summary: The invention describes a system for synchronously aggregating occupational specifications using intelligent neural networking. It utilizes standardized data from the US Department of Labor to create job canon configurations. The system employs bi-directional recurrent neural networks for matching candidates to roles based on recruitment prerequisites.

neural networking, occupational specifications, data aggregation, machine learning

Patent

A model for positron annihilation in multi-layer systems by solving the diffusion equation using different positron affinities

Published on 2026-04-30 by Lucian Mathes, Michael Gldl, Michael Leitner, Bettina Kohlhaas, Maximilian Suhr, Vassily Vadimovitch Burwitz, Armin Manhard and Christoph Hugenschmidt @IOP SCIENCE

Abstract: We present a method for solving the positron diffusion equation in multi-layer systems. Our approach incorporates material-specific implantation profiles, diffusion parameters, and positron affinities. It utilizes a Markov chain approach to model annihilation probabilities and provides fitting capabilities for experimental S (lineshape) parameter data. We have implemented this algorithm in Python and made it available for free under the name layer-wise investigation of measurements on positron i[...]


Our summary: This study presents a model for positron annihilation in multi-layer systems using a diffusion equation. It introduces a Python algorithm called LIMPID for analyzing positron implantation and diffusion. The tool enhances the reproducibility of positron defect characterization measurements across research groups.

positron annihilation, diffusion equation, multi-layer systems, Markov chain

Publication

CFD-based analysis of chimney performance optimization for the Manzanares pilot plant

Published on 2026-03-19 by @OXFORD

Abstract: AbstractThis research conducts an extensive computational fluid dynamics (CFD) analysis aimed at optimizing the geometric configuration of solar chimney power plants (SCPPs), with particular emphasis on the Manzanares pilot plant. The research analyses the effects of chimney diameter and divergent chimney design on system performance under steady-state conditions at 1000 W/m2 solar irradiance and 300 K ambient temperature. A validated three-dimensional CFD model with a 90° symmetry sector is [...]


Our summary: This research utilizes CFD to optimize chimney design for the Manzanares pilot plant. The study finds that increasing chimney diameter significantly boosts power output, while a divergent configuration further enhances performance. Optimal design parameters lead to substantial improvements in airflow and energy generation, providing a cost-effective solution for future installations.

CFD, chimney optimization, solar chimney power plants, aerodynamic design

Publication

Sujets abordés : Calcul haute performance, traitement à vitesse extrême, systèmes informatiques, grappes, superordinateurs, systèmes hétérogènes, CPU, GPU, MPI, OpenMP, parallélisme, gestion des ressources, équilibrage des charges, tolérance aux pannes, surcharge de communication, simulations scientifiques, analyse intensive des données, traitement en temps réel ISO/IEC 30170, IEEE 754, ISO/IEC 13818, IEEE 8023, et ISO/IEC 19785.

Glossaire des termes utilisés

High-Performance Computing (HPC): un paradigme informatique qui permet l'exécution de calculs complexes et le traitement de données à des vitesses et avec une efficacité nettement supérieures à celles de l'informatique traditionnelle, utilisant généralement le traitement parallèle, des clusters à grande échelle et des architectures avancées pour résoudre des problèmes scientifiques, d'ingénierie et analytiques.

Contexte historique

1965
1970
1970
1974-11-15
1980
1980
1980
1964
1968
1970
1970
1975
1980
1980
1980

(si la date est inconnue ou non pertinente, par exemple « mécanique des fluides », une estimation arrondie de son émergence notable est fournie)

Les images en pleine résolution et les téléchargements sont uniquement disponibles, et 100% gratuits, pour les membres inscrits.