Search results for " array"
showing 10 items of 895 documents
Wireless versus Wired Network-on-Chip to Enable the Multi- Tenant Multi-FPGAs in Cloud
2021
The new era of computing is not CPU-centric but enriched with all the heterogeneous computing resources including the reconfigurable fabric. In multi-FPGA architecture, either deployed within a data center or as a standalone model, inter-FPGA communication is crucial. Network-on-chip exhibits a promising performance for the integration of one FPGA. A sustainable communication architecture requires stable performance as the number of applications or users grows. Wireless network-on-chip has the potential to be that communication architecture, as it boasts the same performance capability as wired solutions in addition to its multicast capacities. We conducted an exploratory study to investiga…
Feasibility of FPGA accelerated IPsec on cloud
2018
Abstract Hardware acceleration for famous VPN solution, IPsec, has been widely researched already. Still it is not fully covered and the increasing latency, throughput, and feature requirements need further evaluation. We propose an IPsec accelerator architecture in an FPGA and explain the details that need to be considered for a production ready design. This research considers the IPsec packet processing without IKE to be offloaded on an FPGA in an SDN network. Related work performance rates in 64 byte packet size for throughput is 1–2 Gbps with 0.2 ms latency in software, and 1–4 Gbps with unknown latencies for hardware solutions. Our proposed architecture is capable to host 1000 concurre…
PNeuro: A scalable energy-efficient programmable hardware accelerator for neural networks
2018
Proceedings of a meeting held 19-23 March 2018, Dresden, Germany; International audience; Artificial intelligence and especially Machine Learning recently gained a lot of interest from the industry. Indeed, new generation of neural networks built with a large number of successive computing layers enables a large amount of new applications and services implemented from smart sensors to data centers. These Deep Neural Networks (DNN) can interpret signals to recognize objects or situations to drive decision processes. However, their integration into embedded systems remains challenging due to their high computing needs. This paper presents PNeuro, a scalable energy-efficient hardware accelerat…
Multiple register synchronization with a high-speed serial link using the Aurora protocol
2013
In this work, the development and characterization of a multiple synchronous registers interface communicating with a high-speed serial link and using the Aurora protocol is presented. A detailed description of the developing process and the characterization methods and hardware test benches are also included. This interface will implement the slow control busses of the digitizer cards for the second generation of electronics for the Advanced GAmma Tracking Array (AGATA).
Single-Event Effects in the Peripheral Circuitry of a Commercial Ferroelectric Random Access Memory
2018
International audience; This paper identifies the failure modes of a commercial 130-nm ferroelectric random access memory. The devices were irradiated with heavy-ion and pulsed focused X-ray beams. Various failure modes are observed, which generate characteristic error patterns, affecting isolated bits, words, groups of pages, and sometimes entire regions of the memory array. The underlying mechanisms are discussed.
Optical Link Card Design for the Phase II Upgrade of TileCal Experiment
2011
This paper presents the design of an optical link card developed in the frame of the R&D activities for the phase 2 upgrade of the TileCal experiment. This board, that is part of the evaluation of different technologies for the final choice in the next years, is designed as a mezzanine that can work independently or be plugged in the optical multiplexer board of the TileCal backend electronics. It includes two SNAP 12 optical connectors able to transmit and receive up to 75 Gb/s and one SFP optical connector for lower speeds and compatibility with existing hardware as the read out driver. All processing is done in a Stratix II GX field-programmable gate array (FPGA). Details are given on th…
Search for gamma ray bursts with the ARGO-YBJ detector in scaler mode
2009
We report on the search for Gamma Ray Bursts (GRBs) in the energy range 1-100 GeV in coincidence with the prompt emission detected by satellites using the Astrophysical Radiation with Ground-based Observatory at YangBaJing (ARGO-YBJ) air shower detector. Thanks to its mountain location (Yangbajing, Tibet, P.R. China, 4300 m a.s.l.), active surface (about 6700 m**2 of Resistive Plate Chambers), and large field of view (about 2 sr, limited only by the atmospheric absorption), the ARGO-YBJ air shower detector is particularly suitable for the detection of unpredictable and short duration events such as GRBs. The search is carried out using the "single particle technique", i.e. counting all the …
Pairing-quadrupole interplay in the neutron-deficient tin nuclei: First lifetime measurements of low-lying states in 106,108Sn
2020
The lifetimes of the low-lying excited states 2(+) and 4(+) have been directly measured in the neutron-deficient Sn-106,Sn-108 isotopes. the nuclei were populated via a deep-inelastic reaction and the lifetime measurement was performed employing a differential plunger device. the emitted gamma rays were detected by the AGATA array, while the reaction products were uniquely identified by the VAMOS++ magnetic spectrometer. Large-Scale Shell-Model calculations with realistic forces indicate that, independently of the pairing content of the interaction, the quadrupole force is dominant in the B(E2; 2(1)(+) -> 0(g.s)(+)) values and it describes well the experimental pattern for Sn104-114 ; the B…
A new beamline for energy-dispersive high-resolution PIXE analysis using polycapillary optics
2019
Abstract A new beamline for high energy resolution PIXE measurements is presented. This new setup includes options for both in-air and vacuum measurements. For the high energy resolution transition-edge sensor array, a polycapillary lens is used for detecting low-energy X-rays down to 0.5 keV and to increase the effective solid angle. X-ray transmission of the polycapillary lens was characterized using two calibration standards. The gain obtained by adding a polycapillary lens was 1.6–2.3 at energies between 2.1 keV and 4.5 keV. From 1.04 to 1.74 keV the gain is increased to 2.1–3.0, and at energies 4.9–8.0 keV the gain is between 1.6 and 0.65. The measured gain agreed well with theoretical…
High homogeneity permanent magnet for diamond magnetometry
2020
Abstract Halbach magnets are a source of homogeneous magnetic field in an enclosed volume while keeping stray fields at a minimum. Here, we present the design, construction, and characterization for a stack of two Halbach rings with 10 cm inner diameter providing a homogeneous ( 100 ppm over 1.0 × 1.0 × 0.5 cm 3 ) magnetic field of around 105 mT, which will be used for a diamond based microwave-free widefield imaging setup. The final characterization is performed with a novel fiberized diamond-based sensor on a 3D translation stage documenting the high homogeneity of the constructed Halbach array and its suitability for the proposed use.