High-performance Processing Research Articles

Performance Characterization of Hardware/Software Communication Interfaces in End-to-End Power Management Solutions of High-Performance Computing Processors

Power management (PM) is cumbersome for today’s computing systems. Attainable performance is bounded by the architecture’s computing efficiency and capped in temperature, current, and power. PM is composed of multiple interacting layers. High-level controllers (HLCs) involve application-level policies, operating system agents (OSPMs), and PM governors and interfaces. The application of high-level control decisions is currently delegated to an on-chip power management unit executing tailored PM firmware routines. The complexity of this structure arises from the scale of the interaction, which pervades the whole system architecture. This paper aims to characterize the cost of the communication backbone between high-level OSPM agents and the on-chip power management unit (PMU) in high performance computing (HPC) processors. For this purpose, we target the System Control and Management Interface (SCMI), which is an open standard proposed by Arm. We enhance a fully open-source, end-to-end FPGA-based HW/SW framework to simulate the interaction between a HLC, a HPC system, and a PMU. This includes the application-level PM policies, the drivers of the operating system-directed configuration and power management (OSPM) governor, and the hardware and firmware of the PMU, allowing us to evaluate the impact of the communication backbone on the overall control scheme. With this framework, we first conduct an in-depth latency study of the communication interface across the whole PM hardware (HW) and software (SW) stack. Finally, we studied the impact of latency in terms of the quality of the end-to-end control, showing that the SCMI protocol can sustain reactive power management policies.

Food Waste as Feedstock for Anaerobic Mono-Digestion Process

There is a growing recognition that food waste (FW) comprises a significant amount of unused energy. Indeed, FW shows great potential to produce methane (CH4)-rich biogas via an anaerobic digestion (AD) process. Nevertheless, to ensure high AD process performance, deepening the knowledge of FW characteristics is required. Furthermore, the biogas yield is strongly influenced by several operational parameters. Taking into account the above, in the current study, based on the data in the literature, the physicochemical parameters of FW generated throughout the world are presented and discussed. In addition, the performance profile of the single-stage anaerobic mono-digestion process with the use of FW as a feedstock was investigated. The performed analysis clearly demonstrated that FW is characterized by significant variations in several parameters, such as pH, the total solid (TS) and volatile solid (VS) contents, the volatile solids to total solids ratio (VS/TS), soluble chemical oxygen demand (sCOD), the concentrations of VFAs and ammonium nitrogen (NH4+-N), and the carbon-to-nitrogen ratio (C/N). Moreover, it was shown that the selected operational parameters, such as temperature, pH, the ratio of food waste to inoculum (I) (FW/I), and the organic loading rate (OLR), may have the most significant impact on the performance of the single-stage anaerobic mono-digestion process. In addition, it was found that most of the experimental investigations presented in the literature were conducted on a laboratory scale. Hence, in future research, more effort should be made to determine the biogas yield with the use of full-scale systems. To summarize, it should be clearly highlighted that the analysis presented in this study may have important implications for the management and application of FW as feedstock for an anaerobic mono-digestion process on an industrial scale.

High performance Si-MoS2 heterogeneous embedded DRAM.

Embedded Dynamic RAM (eDRAM) has become a key solution for large-capacity cache in high-performance processors. A heterogeneous two transistor capacitorless eDRAM (2T-eDRAM) that combines silicon and molybdenum disulfide (MoS2) is reported to address the short retention issue in conventional gain cell (GC) eDRAMs meanwhile eliminate the pillar capacitor in one transistor and one capacitor (1T1C) eDRAMs. The MoS2 write transistor with low OFF current (IOFF) enables long data retention, while the Si read transistor offers high drive current and logic compatibility. This combination enhances data retention by 1000 times and sense margin by 100 times respectively compared to full Si and MoS2 counterparts. A three-dimensional (3D) design stacking MoS2 on Si is demonstrated with back-end-of-line (BEOL) process to double integration density. With 6000 s data retention, 35 μA/μm sense margin, 5 ns access speeds, 3D integration and CMOS logic compatibility, this Si-MoS2 eDRAM marks a significant advancement in memory technology.

Study on micro-grinding mechanism and surface quality of hardened 20 vol% SiCp/Al composites

Particle-reinforced aluminum-based silicon carbide composites (SiCp/Al) are widely used in aerospace, rail transit, and other fields due to their excellent comprehensive properties. However, the incorporation of high-brittle SiC particles poses challenges in the high-performance processing of SiCp/Al composites. This study focuses on hardened 20 vol% SiCp/Al composites, establishing two-dimensional finite element cutting models for single SiC particles. Micro-grinding experiments are conducted under various processing parameters. The influence law of surface defects on individual SiC particles relative to tool position and single factor machining parameters was analyzed, revealing the micro-grinding mechanisms affecting surface and subsurface quality. The results show that the interaction among particle damage mechanisms, crack propagation directions, and the restraining effects of the Al matrix leads to diverse types of defects at different cutting positions. Both simulation and experimental observations reveal surface defects such as pits, gullies, and burrs, while subsurface damage primarily results from crack invasion, cavity interstices, and pits due to particle breakage. The experiment achieved a minimum roughness of 0.057 μm. Taking into account the thermal softening of the material and the micro-behavior of the tool-chip interface, optimal conditions for improved surface quality involve a grinding depth of 0.03 mm, a spindle speed of 12,000 rpm, and a relatively low feed rate.

Improving fatigue life of a titanium alloy through coupled electromagnetic treatments

TC11 titanium alloy is widely used in the manufacture of key components such as blades of gas turbine and aero engine because of its high specific strength and good processing performance. In the case of gas turbine or aero engine, the fatigue performance of TC11 will directly determine the life of the turbine or engine, and the surface residual stress generated on the alloy during manufacturing often affects the fatigue life of the material. In this study, a new method of coupled electromagnetic treatment (CEMT) was applied to regulate the surface residual stress of the alloy after manufacturing, so as to improve the fatigue life of the TC11. The results show that after the CEMT, the residual compressive stress in the length direction and width direction increased by 63.7% and 56.0% respectively, the fatigue life of the TC11 is increased by 39.9%. The microstructure analysis shows that after CEMT, the width of fatigue striations is significantly reduced. This paper proposes that CEMT can be used as an effective method to adjust the residual stress of materials and improve the fatigue life of titanium alloys. The research is also relevant for improvement of the fatigue life of other alloy materials.

High Pressure Processing: Pros and Cons

The food industry now uses high pressure processing to increase the shelf life of foods and maintain their nutritional value. This non-thermal preservation technique can successfully inactivate rotting and spoilage-causing bacteria and enzymes without the need for high temperatures during the preservation process. Food is exposed to high pressure, usually in the range of 100 to 600 MPa. High-performance processing keeps food at or slightly above room temperature to prevent heat-sensitive flavors and nutrients from being thermally destroyed during food preparation. This contributes to the preservation of the natural sensory and nutritional properties of the product. Because this process is able to eliminate bacteria and organisms responsible for spoilage, it is a very efficient way to increase the time available to eat foods. Unlike traditional heat-based methods, HPP inactivates harmful microorganisms and enzymes through the application of extreme pressure (100–600 MPa) at low temperatures, thereby reducing thermal degradation. HPP also helps retain bioactive compounds, making it an ideal choice for consumers looking for minimally processed, high-quality foods. This article explores the principles, applications, and benefits of HPP, as well as the challenges and future prospects of its integration into food processing systems.

High-Performance Processing Electronics for Cooler Temperature Sensors in Infrared Payloads

Remote Sensing Infra-Red (IR) imaging sensors operate at very low temperatures (40–100 K). Generally, these types of imaging sensors are integrated with cryo-coolers. These sensors have an inbuilt 2N2222 diode-based cooler temperature sensor (CTS). This paper presents details of low-noise, high=precision processing electronics developed for CTS. Critical design aspects, simulation and hardware results of the system are discussed in this paper. Developed electronics systems achieve ≈5mK resolution and the power dissipation recorded is < 200 mW.

Design of Low Power Control Unit for RISC-V Processor Core

This research work focuses on the development of a low-power decode logic for a RISC-V processor core with specifications. The goal is to create a controller that performs all six groups of instruction formats outlined in the RV32I Base Integer Instruction Set. The control unit is designed to decode a total of 13 instruction sets, allowing for a comprehensive range of operations. A single instruction pipeline approach is implemented in the design to optimize performance. The synthesis of the design is carried out using the 32 nm standard library, resulting in a maximum operating frequency of 666.67 MHz. To further enhance power efficiency, clock gating techniques are employed, leading to a reduction in power consumption by 18.72 % from 112.15 µW to 91.45 µW. Additionally, the layout of the design is optimized, resulting in an area of 354.74 mm2. The successful development of this low-power decode logic demonstrates its potential for integration into larger RISC-V processor cores. Future enhancements can include expanding the instruction decoding capability to encompass the full range of 47 instructions in the RV32I Base Integer Instruction Set, as well as exploring additional pipeline stages to further improve performance. The results achieved in this project contribute to the ongoing pursuit of power-efficient and high-performance processor designs based on the RISC-V architecture.

Efficient Privacy-Preserving Machine Learning with Lightweight Trusted Hardware

In this paper, we propose a new secure machine learning inference platform assisted by a small dedicated security processor, which will be easier to protect and deploy compared to today's TEEs integrated into high-performance processors. Our platform provides three main advantages over the state-of-the-art: (i) We achieve significant performance improvements compared to state-of-the-art distributed Privacy-Preserving Machine Learning (PPML) protocols, with only a small security processor that is comparable to a discrete security chip such as the Trusted Platform Module (TPM) or on-chip security subsystems in SoCs similar to the Apple enclave processor. In the semi-honest setting with WAN/GPU, our scheme is 4X-63X faster than Falcon (PoPETs'21) and AriaNN (PoPETs'22) and 3.8X-12X more communication efficient. We achieve even higher performance improvements in the malicious setting. (ii) Our platform guarantees security with abort against malicious adversaries under honest majority assumption. (iii) Our technique is not limited by the size of secure memory in a TEE and can support high-capacity modern neural networks like ResNet18 and Transformer. While previous work investigated the use of high-performance TEEs in PPML, this work represents the first to show that even tiny secure hardware with very limited performance can be leveraged to significantly speed-up distributed PPML protocols if the protocol can be carefully designed for lightweight trusted hardware.

Design of an active wing-folding biomimetic flapping-wing air vehicle

In nature, birds and bats dynamically alter their wing shapes to suit various flight environments and tasks. This paper focuses on the design and validation of a biomimetic flapping-wing aerial vehicle, named FlexiWing, which features a unique mechanism for active wing deformation. This mechanism allows the wings to adjust their shapes flexibly in response to flight demands, significantly enhancing attitude control and maneuverability.’ ‘This study began with an in-depth exploration of biomimetic principles, focusing particularly on how birds and bats achieve precise control during flight through active wing deformation. Subsequently, we present a detailed account of the design and fabrication process of the active folding biomimetic flapping-wing aerial vehicle, including the design of mechanical mechanisms and material selection. Utilizing lightweight nylon materials and hollow carbon fiber rods, we successfully constructed a mechanically foldable wing structure. To achieve precise control over the aircraft’s movement, an embedded control system was designed, comprising an onboard embedded flight controller and ground-based equipment. The onboard controller uses a high-performance ESP32-C3 processor and a JY901 inertial measurement unit to acquire real-time attitude information of the aircraft. The control system incorporates Wi-Fi communication technology, enabling operators to send commands via a remote control or personal computer to manage flight modes and attitudes. Ultimately, a series of flight experiments were conducted to validate the performance of FlexiWing. The results demonstrate that FlexiWing exhibits remarkable maneuverability and stability, capable of achieving high-precision attitude control through active wing folding, making it adaptable to complex environments and tasks.’

Green Computing: Crafting A Low-Energy Amba APB Bridge Solution

Abstract—The Advanced High-performance Bus (AHB), a key component of the Advanced Microcontroller Bus Architecture (AMBA) family, stands as a high-performance, low-power, and high- bandwidth communication interface within system components. Bridging to low-bandwidth peripherals, the Advanced Peripheral Bus (APB) offers a simple, non-pipelined protocol for read- and-write communication, linking a bridge/master to numerous slaves via a shared bus. The AHB to APB bridge serves a critical role in integrating diverse bus protocols, facilitating efficient communication between high-performance processing units and slower peripherals, thereby enhancing the System-on-Chip (SoC)'s overall effectiveness and functionality. This approach addresses the functional verification of the AMBA AHB to APB Bridge protocol with a focus on completeness. We employ a layered testbench architecture in System Verilog to ensure thorough verification of functionality with maximal coverage. Our verification environment is constructed using System Verilog, and simulations are conducted on the EDA playground platform. Through this comprehensive approach, we aim to provide robust validation of the AHB to APB bridge protocol, ensuring its reliability and compatibility within complex SoC designs. Keywords— Advanced Peripheral Bus (APB), System-On-Chip (SOC), Advanced Microcontroller Bus Architecture.

Adaptive data quality scoring operations framework using drift-aware mechanism for industrial applications

Within data-driven artificial intelligence (AI) systems for industrial applications, ensuring the reliability of the incoming data streams is an integral part of trustworthy decision-making. An approach to assess data validity is data quality scoring, which assigns a score to each data point or stream based on various quality dimensions. However, certain dimensions exhibit dynamic qualities, which require adaptation on the basis of the system’s current conditions. Existing methods often overlook this aspect, making them inefficient in dynamic production environments. In this paper, we introduce the Adaptive Data Quality Scoring Operations Framework, a novel framework developed to address the challenges posed by dynamic quality dimensions in industrial data streams. The framework introduces an innovative approach by integrating a dynamic change detector mechanism that actively monitors and adapts to changes in data quality, ensuring the relevance of quality scores. We evaluate the proposed framework performance in a real-world industrial use case. The experimental results reveal high predictive performance and efficient processing time, highlighting its effectiveness in practical quality-driven AI applications.

1D-CNN-Transformer for Radar Emitter Identification and Implemented on FPGA

Deep learning has brought great development to radar emitter identification technology. In addition, specific emitter identification (SEI), as a branch of radar emitter identification, has also benefited from it. However, the complexity of most deep learning algorithms makes it difficult to adapt to the requirements of the low power consumption and high-performance processing of SEI on embedded devices, so this article proposes solutions from the aspects of software and hardware. From the software side, we design a Transformer variant network, lightweight convolutional Transformer (LW-CT) that supports parameter sharing. Then, we cascade convolutional neural networks (CNNs) and the LW-CT to construct a one-dimensional-CNN-Transformer(1D-CNN-Transformer) lightweight neural network model that can capture the long-range dependencies of radar emitter signals and extract signal spatial domain features meanwhile. In terms of hardware, we design a low-power neural network accelerator based on an FPGA to complete the real-time recognition of radar emitter signals. The accelerator not only designs high-efficiency computing engines for the network, but also devises a reconfigurable buffer called “Ping-pong CBUF” and two-level pipeline architecture for the convolution layer for alleviating the bottleneck caused by the off-chip storage access bandwidth. Experimental results show that the algorithm can achieve a high recognition performance of SEI with a low calculation overhead. In addition, the hardware acceleration platform not only perfectly meets the requirements of the radar emitter recognition system for low power consumption and high-performance processing, but also outperforms the accelerators in other papers in terms of the energy efficiency ratio of Transformer layer processing.

Corrosion protective properties on 2024‐T3 of chitosan‐modified montmorillonite epoxy coatings based on Ce3+ loading

Abstract2024 aluminum alloy has high strength, processing performance, and good damage tolerance performance, but it is prone to intergranular corrosion and pitting. Therefore, the development of corrosion protective coatings is essential for the aluminum alloy. In this paper, cerium ion (Ce3+) was doped into chitosan‐modified montmorillonite (Ce‐CS‐MMT) by ion exchange method, and the corrosion protective performance of Ce‐CS‐MMT epoxy coating (Ce‐CS‐MMT/EP) was investigated. The electrochemical analysis of aluminum alloy in cerium‐containing solution showed the corrosion inhibition efficiency reached 95.52%. The result indicates the corrosion inhibition effect of cerium‐containing solution was better than pure NaCl solution and MMT solution. The corrosion study of the coating illustrates that the protection rate of Ce‐CS‐MMT/EP reached 95.82%, which was higher than that 91.83% of Ce‐MMT/EP. The result indicates the resistance of charge transfer was greater and the better shielding effect for electrolyte corrosion ions than MMT and Ce‐MMT epoxy coatings, and the corrosion protective performance was excellent.

High Performance Receiving and Processing Technology in Satellite Beam Hopping Communication

High Performance Receiving and Processing Technology in Satellite Beam Hopping Communication

Forming limit diagram of annealed copper OFE thick sheets for optimized hydroforming of superconducting RF cavities

Coated superconducting radio-frequency (SRF) cavities are planned to be implemented in the Future Circular Collider (FCC) at the European Organization for Nuclear Research (CERN). The interest in industrialized and high-performance processes for fabrication of the corresponding copper substrates has thus risen. Tubular hydroforming is a well-known industrial process, but has not been thoroughly applied for ultra-pure oxygen-free electronic copper (Cu-OFE) thick products. To provide a feasibility analysis for the optimized fabrication process of copper substrate seamless cavities, a Forming Limit Diagram (FLD) of the material of interest was studied. In this paper, the methodology, test results and initial benchmark of the material model is presented, for 4 mm and 2 mm thick sheets.

Synergistic improvement of strength and ductility by high magnetic field assisted intercritical annealing in lightweight medium Mn steel

The microstructures and mechanical properties of lightweight medium Mn steels were studied by high magnetic field (HMF) assisted intercritical annealing process (HMF-IA). The results show that the volume fraction of retained austenite increased and reached the maximum value of 56.2 % when the magnetic flux density (MFD) was 10 T at first and then decreased with the increase of MFD. The application of 10 T HMF promoted the diffusion ability of carbon atoms into austenite by magnetic stress-driven dislocation movement and increased the volume fraction and grain size of austenite. HMF also played a critical role in dislocation movement and element partitioning during the IA process. Due to the wide range of bimodal size distribution of austenite after the HMF-IA process, stronger transformation induced plasticity effect and twin induced plasticity effect happened in the process of plastic deformation. For the sample annealed under 10 T, comparing with the sample annealed without HMF, the product of strength and elongation value increased from 45.7 GPa·% to 55.1 GPa·%. Additionally, it also kept a persistent high strain hardening. The present study provides a new understanding of the role of HMF in adjusting the relationship between microstructure and mechanical properties during the annealing process and opens up a new path for designing industrial medium Mn steel with low cost, high performance and simple process.

Game psychotherapy intervention based on entertainment interactive robots for preventing depression in university students

This study designs an entertainment interactive robot system for gaming psychotherapy intervention and evaluates its effectiveness in preventing depression in students. In terms of the architecture and hardware of the entertainment interactive robot system, the robot adopts high-performance processors and sensors to achieve the ability to quickly respond and accurately perceive user actions. In terms of software, a mathematical model was established based on the structure and kinematic principles of the robot to describe its motion. Preprocess user gesture data by analyzing the features and patterns of user gestures, extracting key action information, and using machine learning algorithms or deep learning models to classify and recognize gestures to determine user intent and action types. Evaluate action recognition and effectiveness, and verify the accuracy and reliability of the gesture interaction module through comparison and testing with actual user gestures. The study designed an entertainment interactive robot assisted game that provides a fun and interactive experience to attract students’ attention and stimulate positive emotions. Through game design and behavioral pattern evaluation, the research team evaluated the preventive effect of entertainment interactive robot assisted games on depression and proposed optimization strategies.

Modern approaches to data storage: comparison of relational and cloud data warehouses using etl and elt methods

The paper analyses various aspects of the use of relational and cloud data warehouses as well as methods of integrating ETL and ELT data. A comparative analysis of these approaches, their advantages and disadvantages are provided. A central relational data warehouse is proposed that provides a single version of truth (SVOT), which allows standardising and structuring data, avoiding differences and providing the access to the same information for all users of an organisation. It is analysed the methodological approaches to implementing a data warehouse: top-down, bottom-up, and from middle. It is described cloud data warehouses that use cloud technologies to provide scalability, availability and fault tolerance, which is important for the companies with huge amounts of data. The advantages and disadvantages of ETL and ELT are analysed: ETL transforms data before it is loaded into the warehouse, which makes it easier to maintain data confidentiality. ELT performs transformation after loading, which allows for more flexible data processing directly in the warehouse. In the article, we deal with the approaches to implementing a data warehouse: top-down is suitable for strategic planning, bottom-up allows for faster results, and the middle approach combines both methods to achieve optimal efficiency. We considered cloud data storage: compared to relational storage, cloud storage is more flexible, scalable and efficient, providing speed and reducing infrastructure costs. It is described cloud storage architectures: massive parallel processing, hybrid architectures, lambda architectures, and multi-structured architectures. They provide high performance and flexibility in data processing. It is described data storage technologies: Data Lake, Polyglot Persistence, Apache Iceberg, Apache Parquet, and columnar databases that provide efficient storage and processing of large amounts of data

The effects of electrolyte composition and deposition voltage on the copper-nickel alloy micropillars fabricated by Jet ECD

With the rapid development of micro electronic component technology, higher requirements are put forward for high performance and high precision manufacturing process. In this paper, copper-nickel alloy micropillars were manufactured by jet electrochemical deposition (Jet ECD) technology with high deposition rate, aiming to provide an effective material and process solution for the manufacture of micro-thermocouples and other components. By optimizing the process parameters such as electrolyte composition, pH value and deposition voltage, the homogeneous deposition of copper-nickel alloy micropillar was realized. The cross-section morphology and elementary composition of the deposition micropillars were characterized in detail by using scanning electron microscopy (SEM) and energy dispersive spectroscopy (EDS). The results show that the adjusted formula and process parameters can stably manufacture copper-nickel alloy micropillars with a nickel-copper ratio (close to 0.8), and compact, void free microstructure.