Search Results (1,525)

Branch prediction errors can reduce superscalar throughput by more than the error frequency alone suggests because a single misprediction can trigger redirect, squash, refetch, refill, and window recovery, which collectively disrupt sustained instruction-level parallelism. This paper presents a quantitative framework that relates prediction accuracy to realized parallelism loss in out-of-order superscalar processors. The framework separates prediction-error frequency, effective recovery cost, and unrealized issue capacity using prediction accuracy (Acc), misprediction rate (MR), effective branch penalty in cycles per misprediction (BP), parallelism loss ratio (PLR), the ratio-based branch sensitivity factor $BSF=PLR/MR$, and the slope-based branch sensitivity factor $\mathit{S-BSF}=\mathit{\partial }PLR/\mathit{\partial }MR$. BSF measures how strongly a particular processor configuration and workload convert prediction errors into lost issue capacity, whereas S-BSF provides a more stable sensitivity estimate when MR approaches zero. The framework is evaluated using timing-detailed gem5 O3 simulations on RV64GC workloads. The evaluation includes controlled branch microbenchmarks and six GAPBS graph workloads, allowing the proposed metrics to be examined under both mechanism-isolating and complex workload conditions. Two complementary controlled sweeps are used. At a fixed processor structure, predictor family and predictor level are varied to determine whether changing the predictor strengthens or weakens the relationship between MR and IPC/PLR. At a fixed predictor configuration, issue width and an effective front-end-depth proxy are varied to measure how the microarchitecture amplifies the performance cost of the remaining prediction errors. Thus, issue width is treated as an amplification variable for branch-prediction failures rather than as an independent performance topic. At the fixed structural point, Tournament and BiMode predictors show strong monotonic MR–PLR relationships on the high-branch benchmark, with Spearman coefficients of 1.00 and 0.98, whereas the Local predictor exhibits nearly unchanged MR but materially different IPC and PLR across levels. This demonstrates that the mapping from MR to throughput depends on predictor family and configuration rather than being invariant. In the controlled structural sweep, increasing issue width from 4 to 8 raises PLR by 37.6% and BSF by 55.0% on the high-branch benchmark, even though MR remains in the same order of magnitude. On GAPBS workloads, the lowest-MR configuration is not always the highest-IPC configuration, confirming that effective branch penalty and parallelism loss must be considered together with prediction frequency. These numerical findings are conditional on the evaluated single-thread gem5 DerivO3CPU model, RV64GC binaries, predictor implementations, memory hierarchy, and workload set. They characterize predictor–microarchitecture interactions in this controlled simulation environment and should not be interpreted as universal constants for all processors or applications. Full article

Background/Objectives: This study aims to reveal the physiological and molecular regulatory mechanisms of the genuine medicinal herb I. suzhouensis K. F. Zhai, Z. B. Han & S. B. Zhou (Wangzaozi) of Anhui province in response to drought stress, and clarify the regulatory patterns of drought adversity on the accumulation of its medicinal active ingredients. Methods: Mild natural drought treatment was applied to I. suzhouensis. Combined with Illumina high-throughput transcriptome sequencing and HPLC-MS/MS targeted metabolomics detection, this study jointly deciphered the dynamic changes in gene expression and metabolite accumulation of I. suzhouensis under drought. Key drought-responsive metabolic pathways, core regulatory genes and marker metabolites were screened. Results: A total of 56,823 high-quality unigenes were obtained via transcriptome sequencing, among which 23,580 differentially expressed genes (DEGs) were identified. Functional enrichment analysis revealed that DEGs were predominantly enriched in pathways, including plant hormone signal transduction, phenylpropanoid biosynthesis, flavonoid biosynthesis and photosynthesis. A total of 4171 metabolites were qualitatively and quantitatively characterized via metabolomics, and 1632 differentially expressed metabolites (DEMs) were screened, mainly enriched in phenylpropanoid biosynthesis, tyrosine metabolism, flavone and flavonol biosynthesis pathways. Physiological measurements of antioxidant indices demonstrated that the activities of SOD and POD increased by approximately 2-fold, while PAL activity rose by 1.55-fold, and chlorophyll content decreased significantly. Multi-omics joint analysis indicated that mild drought stress modulates the expression of genes involved in phenylpropanoid, flavonoid and diterpenoid biosynthetic pathways, alters antioxidant enzyme activities, and coordinately regulates the formation of drought tolerance and the accumulation of bioactive compounds in I. suzhouensis. Conclusions: This study systematically elucidates the drought response mechanism of I. suzhouensis cultivated in northern Anhui province. It provides theoretical evidence and candidate core responsive gene resources for standardized cultivation of I. suzhouensis and precise regulation of medicinal quality in drought-prone production areas. Full article

Rhizosphere microorganisms have a significant impact on plant growth and development. To clarify the regulatory mechanisms underlying maize genotype effects on rhizosphere soil physicochemical factors and microbial communities in the Hexi Corridor of China, this study used seven maize varieties (Wugu 305, Xianyu 335, Zhengdan 958, Jingke 968, Yufeng 303, Dedan 1104, and Qinfeng 876) as experimental materials. Rhizosphere soil physico-chemical indicators were measured, and high-throughput sequencing was used to analyze the diversity, structure, unique bacterial communities, and metabolic functions of rhizosphere fungi and bacteria. Correlation analysis was also conducted using soil environmental factors. The results showed that the pH of the tested maize rhizosphere soil was weakly alkaline (7.96–8.10); significant varietal differences were observed in total nitrogen, available nitrogen, and available phosphorus, while no significant differences were observed in organic matter and total phosphorus among maize varieties. Dedan 1104 and Jingke 968 had the highest fungal richness, diversity, and evenness at the phylum and genus level, according to an alpha-diversity study of maize rhizosphere soil microorganisms, which revealed that variation did not affect fungal community richness. Zhengdan 958 has the highest microbial richness at the bacterial phylum level. At the bacterial genus level, Dedan 1104 had the highest microbial abundance. Ascomycota is the predominant fungal phylum, with Proteobacteria, Acidobacteria, and Chloroflexi being the core bacterial phyla. Venn analysis confirmed strong conservation of fungal community composition across various maize varieties; abundant bacterial-specific taxa and greater genotype-driven differentiation. Furthermore, significant differences were found in the enriched microbial communities in the rhizosphere among different maize genotypes, with fungal metabolic functions strongly influenced by maize genotype, and the distribution of core bacterial functional enzymes tending towards homogeneity. Available phosphorus and alkaline-available nitrogen were the core environmental factors regulating the structure of rhizosphere fungal and bacterial communities, and the dominant functional microorganisms were significantly positively correlated with these nutrients. This study provides a theoretical basis for screening high-quality seed maize varieties and regulating rhizosphere soil microecology. Full article

Cereal crops, including wheat and barley, are essential for global food security, but their productivity is strongly affected by nitrogen availability and water limitation. This study investigated the phenotypic responses of two commercially significant spring wheat cultivars, Videodur (DU) and Sensas (SW), and two spring barley cultivars, Tiroler Imperial (SG1) and Amidala (SG2), exposed to two nitrogen regimes, low nitrogen at 25 kg N/ha (N25) and high nitrogen at 130 kg N/ha (N130), under drought and well-watered conditions. Plants were monitored from the late vegetative stage through maturity under controlled multivariable climatic conditions similar to field settings. A high-throughput phenotyping workflow was applied, combining precision watering, RGB imaging, infrared thermography, and VNIR–SWIR hyperspectral imaging to quantify plant growth, projected digital biomass, plant temperature, water use efficiency, and spectral vegetation indices associated with pigment dynamics, water status, maturation, and senescence. The results revealed cultivar-specific responses to combined nitrogen and drought stress. Under drought conditions, the high nitrogen treatment (N130) increased plant temperature (Tplant) for barley (cv. SG1) and wheat (cv. SW) compared to N25, thereby accelerating early maturation. However, the decline in chlorophyll was not uniformly faster across all cultivars tested. The DU cultivar exhibited superior chlorophyll absorption and reflectance, indicating better drought adaptation compared to other tested species. The high nitrogen treatment (N130) reduced water use efficiency (WUE) in the SW and SG2 cultivars compared to N25, implying that these cultivars used more water. Enhanced nitrogen did not consistently improve water use efficiency but did accelerate the growth cycle. SG2 was particularly sensitive to drought, showing declines in vegetation indices, except for the Water Content Index, highlighting the need for precise water and nitrogen management. Overall, the integration of hyperspectral, thermal, RGB, and water use measurements enabled the identification of trait signatures linked to drought adaptation, nitrogen response, maturation, and senescence. These findings provide practical insights for optimizing nitrogen and irrigation management and for supporting breeding strategies aimed at improving cereal crop resilience under climate-change-associated stress conditions. Full article

Curdlan is an industrially important β-1,3-glucan with applications in the food, pharmaceutical, and biomaterial industries. However, the identification of high-yielding curdlan-producing strains is hindered by the absence of rapid and efficient screening methods. To address this limitation, we developed a systematically optimized integrated microscale workflow combining 48-well plate fermentation with a quantitative aniline blue-based colorimetric assay in 96-well plates for high-throughput screening of curdlan-producing strains. Fermentation was miniaturized using 48-well plates, while curdlan quantification was performed in 96-well plates through formation of a curdlan–aniline blue complex. Key parameters were systematically optimized. Under optimal conditions, curdlan dissolved in 1.0 mol/L NaOH was reacted with 2.0 mg/mL aniline blue in 0.5 mol/L phosphate buffer (pH 7.0) for 90 min, and absorbance was measured at 550 nm. The assay demonstrated excellent linearity between curdlan concentration and absorbance (y = 1.0008x + 0.1536, R² = 0.9957, p < 0.001). To validate the method, curdlan yields from nine mutants derived from ATCC31749 were determined using both gravimetric and colorimetric approaches, revealing a strong correlation (R² = 0.8683, p < 0.001), that confirmed the assay’s reliability for rapid screening. Application of this platform to 132 UV-mutagenized strains identified nine mutants with enhanced curdlan production. The best-performing strain, UV150824-02, produced 46.8 ± 0.08 g/L curdlan, an 11.4% increase over the wild-type strain ATCC31749 (41.6 ± 1.54 g/L), and maintained stable curdlan production over nine laboratory passages (coefficient of variation = 2.72%). This method significantly improves the efficiency of mutagenesis-based strain screening and provides a practical and efficient tool for accelerating strain improvement in industrial polysaccharide fermentation. Full article

Smart cities (SCs) use the Internet of Things (IoT) to collect and process data to communicate with their infrastructure and assets in real time. A great deal of techniques, such as encryption protocols, Random Forest-based AI-driven threat detection, and blockchain architectures, have been developed to address cybersecurity challenges in smart cities (SCs). These techniques, however, have limitations such as their scalability, high computational expenses, and energy inefficiency. Therefore, in this study, to overcome these challenges, we propose a blockchain-based infrastructure called BlockSafeNet. This uses artificial intelligence, big data, and blockchain to enhance cybersecurity in SCs. The effectiveness of the proposed BlockSafeNet framework was evaluated using responsiveness, computational time, encryption quality score, detection rate, false positive rate, latency, throughput, and energy consumption as the primary cybersecurity performance metrics. These metrics were selected to assess communication efficiency, threat detection capability, privacy preservation, scalability, and overall security performance within smart-city IoT environments. To ensure secure data transactions, robust threat detection, and efficient communication. The system’s high calculation speed and detection rate show potential for managing sensitive maternal health data collected by IoT devices. The platform also shows how IoT may be used by healthcare services to monitor public health in real time, allowing hospitals, emergency services, and public health agencies to securely share data. This aids in resource optimization, improving service delivery, and preserving data privacy and trust in SCs. Data was obtained from the UCI Machine Learning Repository on Kaggle to validate the developed framework. By evaluating the effectiveness of BlockSafeNet in tackling cybersecurity challenges, we establish its practical relevance and usability in SCs. The proposed BlockSafeNet framework achieved a responsiveness of 24 s, an encryption quality score of 0.89, computational time of 85 s, and a detection rate of 91%, demonstrating significant improvements in secure IoT communication, privacy preservation, and AI-driven cyber threat detection within smart city infrastructures. shows that SC IoT security has significantly improved through the adoption of new data protection methods and better measures of security, providing a positive impact on the SC ecosystem. Full article

Photovoltaic-integrated smart grids require energy storage systems that mitigate net load fluctuations without imposing excessive operating burden on battery subsystems. This study formulates a stress-prioritized hierarchical MPC-based control allocation strategy for a lead-acid/LiFePO₄ hybrid energy storage system. Unlike grid-smoothing-oriented controllers, the proposed method explicitly incorporates lead-acid operating stress into the control objective and evaluation framework. The upper layer schedules the lead-acid battery using a low-frequency net load component while penalizing power magnitude, ramping, throughput, and state-of-charge deviation. The lower layer controls the LiFePO₄ battery to compensate residual net load variations and reduce the burden imposed on the lead-acid subsystem. The method was evaluated using 696 hourly samples of measured photovoltaic generation and load demand data from the Naju Sports Park smart-grid site. Compared with the lead-acid-only MPC case, the proposed strategy reduced lead-acid throughput and equivalent full cycles by 66.5%, ramp burden by 76.4%, high-power operation time by 78.8%, and high-power operation energy by 81.0%. Compared with rule-based hybrid control, it reduced lead-acid throughput and equivalent full cycles by 38.4% while accepting a 3.3% increase in grid power standard deviation. These results indicate that the proposed strategy provides a practical stress-prioritized operating framework for lead-acid/LiFePO₄ hybrid energy storage systems. Full article

Edge artificial-intelligence (AI) inference energy comparisons can mislead when timing, system-energy integration, and requested hardware states are conflated. We audited 558 Jetson Orin Nano runs; the balanced primary design comprised 378 runs on three distinct physical boards, spanning two convolutional neural networks (CNNs), FP16/FP32/FP64 tensor dtypes, seven batch sizes, and three requested profile branches. Forward performance was separated from onboard input-rail (VDD_IN) energy over the logged window. In the full grid, precision accounted for 71.5% of log-scale forward-throughput variation and 66.4% of logged-window energy variation, largely reflecting FP64 stress. In the FP16/FP32 slow + medium sensitivity, batch instead accounted for 54.1% of forward-throughput variation. Across these cross-board-stable branches, FP32 retained 68.4% and 54.8% of FP16 throughput for MobileNetV2 and ResNet-50, while using 1.212 and 1.612 times the logged energy. At batch one, FP16 retained only 85.4% and 84.7% of FP32 throughput. All 126 requested-fast runs exhibited an exact observed association: the realized low/high clock regime matched whether the last non-fast predecessor was slow (57/57) or medium (69/69). Thus, requested profile labels did not define independently realized treatments. Reliable claims require compatible measurement boundaries, explicit protocol overheads, and verified state reset/read-back. Full article

In semiconductor and optoelectronics manufacturing, the reliability of turbo molecular pumps (TMPs) is vital for maintaining vacuum integrity and ensuring product yield. However, acoustic monitoring in cleanrooms faces severe challenges due to ambient noise levels routinely exceeding 80 dBA and the scarcity of labeled anomaly data. This study proposes an unsupervised acoustic anomaly detection and localization system to address these issues. The performance of the proposed framework is evaluated using an acoustic dataset collected from operational turbopumps in an industrial semiconductor cleanroom environment, encompassing both normal operations and naturally occurring failure conditions. We introduce a Convolutional Autoencoder (CAE) based on ResNet-18, integrated with a Convolutional Block Attention Module (CBAM) to adaptively suppress high-decibel environmental noise. To enhance sensitivity to structural spectral defects, a hybrid loss function combining Mean Squared Error (MSE) and structural similarity index measure (SSIM) is implemented. Experimental results, supported by rigorous hyperparameter sensitivity analysis, demonstrate that the proposed model achieves an outstanding AUC of 0.9535 and a fault recall of 99.06% under a validation-calibrated threshold, significantly outperforming standard U-Net architectures. Furthermore, the system generates anomaly heatmaps for precise time–frequency localization, enabling explainable diagnostics. With a model inference throughput of 330.28 FPS and an end-to-end processing rate of ≈73× in real time, the proposed framework provides an efficient and robust solution for real-time predictive maintenance in noisy industrial settings. Full article

Compared to conventional signalised intersections, roundabouts are increasingly recognised for their ability to improve traffic safety and operational efficiency. However, accurately modelling their complex traffic dynamics remains challenging, particularly in multilane configurations characterised by lane-changing manoeuvres and gap-acceptance interactions. This study presents a behaviour-driven microscopic simulation framework based on agent-based modelling (ABM) for evaluating roundabout performance under varying geometric and traffic demand conditions. In the proposed framework, each vehicle is represented as an autonomous agent capable of route selection, yielding, lane-changing, and speed adaptation according to predefined behavioural rules. This enables a detailed representation of local traffic interactions and operational conflicts that are not fully captured by traditional aggregate traffic models. The simulation environment is used to analyse idealised one-, two-, and three-lane roundabout configurations and to assess the operational impact of targeted geometric modifications. The proposed methodology is further validated using real-world traffic data collected from the Brașov Central Roundabout, Romania. Simulation results demonstrate that the ABM framework can realistically reproduce traffic throughput, average speed, number of stops, and travel time under high traffic demand conditions. Furthermore, the introduction of a channelised right-turn lane resulted in measurable operational improvements, including increased average speed and reduced delay. The findings highlight the applicability of agent-based simulation as a decision-support tool for roundabout design, traffic management, and infrastructure optimisation, contributing to safer and more efficient urban mobility systems. Full article

Arbuscular mycorrhizal fungi (AMF) are the key symbiotic microorganisms in sandy poplar plantations, regulating vegetation restoration and nutrient cycling. This study aims to investigate the impact of biogas slurry (BS) application on the community structure of AMF in the rhizosphere soil of poplar forests, in order to support the sustainable management of the forest. This study focused on poplar plantations in the sandy areas of Northeast China. High-throughput sequencing technology was utilized to examine the community structure and diversity characteristics of AMF while measuring the physical and chemical characteristics of the soil. The application of BS increased the nutritional status of poplar rhizosphere soil and changed its physical and chemical characteristics, according to the findings. Paraglomus and Glomus dominated the AMF community. The application of BS increased the relative abundance of Glomus in the 0–20 cm and 20–40 cm soil strata. The primary determinants of changes in the dominant AMF populations of poplar were TP, AP, TN, NO₃⁻-N, and pH. To conclude, BS application together with soil depth regulates AMF community composition within sandy poplar plantations. BS amendment ameliorates soil nutrient conditions, increasing subsurface AMF diversity and the relative abundance of vital functional genera. Three core hypotheses were tested: (1) BS application elevates soil carbon, nitrogen and phosphorus concentrations. (2) Biogas slurry and soil depth jointly alter AMF richness and composition. (3) BS-induced shifts in soil nutrients are significantly correlated with AMF community assemblage. Full article

Growth in portable electronics and electric vehicles has increased the diversity of battery types and chemistries, intensifying demands for fast and accurate assessment within End-of-Life recycling systems. Conventional sorting and disassembly rely heavily on manual or slow diagnostics, poorly suited to high-throughput recycling environments. This study presents an Automated Machine Learning (AutoML) framework for non-destructive battery cell chemistry classification using short-measurement data from a publicly available dataset comprising 109 cells spanning six battery chemistries. The data include 20 raw descriptors capturing physical properties, DC load-response behaviour, and impedance measurements, from which three ageing-informed features are engineered. The framework integrates leakage-safe preprocessing, Minimum Redundancy Maximum Relevance (MRMR) feature selection, and automated learner optimisation within a modelling pipeline. Six experimental cases benchmark deterministic and AutoML-selected classifiers, assess feature representations, identify a compact MRMR Top-4 subset, and evaluate robustness under repeated resampling and constrained training availability. The best configuration achieves 92.9% accuracy and Macro-F1 of 0.923 under deterministic evaluation, and a mean accuracy of 97.6% and Macro-F1 of 0.964 with narrow empirical variability intervals under repeated cross-validation. Overall, within the scope of the dataset examined, the proposed workflow provides an interpretable and reproducible methodological foundation for battery sorting and recycling applications. Full article

Embedded systems increasingly require on-device deep learning, yet their processors must simultaneously handle real-time sensing, networking administration, and data control. Existing Field-Programmable Gate Array (FPGA) accelerators typically target peak per-operator throughput without addressing concurrent execution demands of real-time embedded platforms. This paper presents a systolic array-based accelerator prototype implemented on the Zynq-7000 SoC integrated directly into PyTorch, enabling dense linear algebra to be delegated to the FPGA chip while Cortex-A9 continues executing the software stack uninterrupted. Unlike traditional accelerators optimized for peak per-operator speed, this design prioritizes asynchronous co-executionbetween the processing system (PS, the dual-core Cortex-A9) and the programmable logic (PL): The PL performs tiled matrix multiplication, while the PS executes preprocessing, orchestration, and I/O data concurrently, increasing effective end-to-end throughput regardless of the relative isolated performance of CPU and FPGA. The proposed module includes high-level-synthesis (HLS)-based matrix multiplication, activation functions, and Advanced eXtensible Interface (AXI)-Stream Direct Memory Access (DMA) interfaces, wrapped as custom PyTorch kernels under the PetaLinux operating system. The results obtained on the PYNQ-Z2 board show that, once the DMA transfer time is included in the measurement, the FPGA path does not surpass Cortex-A9 in isolated per-operator latencies across the evaluated range; the benefit lies instead in delegating the matrix compute to the fabric at low incremental power while the host CPU cores stay available for concurrent tasks. A concurrent workload sweep across matrix sizes from $8\times 8$ to $256\times 256$ confirms that the co-execution mode sustains 98–99% of available PS compute throughput compared with a constant ≈50% in single-core blocking mode; the difference is statistically significant for all evaluated sizes (see Mann–Whitney U: $U=25$, $p=3.97\times {10}^{-3}$, perfect discrimination, $n=5$). A fair dual-core CPU-only baseline attains comparable PS availability, so this figure reflects the dual-core scheduling that co-execution enables rather than a per-operator advantage of the fabric; the accelerator’s distinct role is to perform the matrix arithmetic off the general-purpose cores at low incremental power. The design occupies only 8% of available look-up tables (LUTs) and 5% of digital signal processing (DSP) blocks, maintains 1.69 W power with a junction temperature of 44.5 °C, and achieves 96.10% MNIST accuracy under fixed-point arithmetic. Full article

High-channel-count neural recording requires wireless links with high throughput, low power, and compact implementation, yet commercial protocols and phase-locked loop (PLL)-based transmitters often trade data rate against power and complexity. We present a low-power, PLL-less wideband on–off keying (OOK) neural-signal transmitter fabricated in a 180 nm CMOS process. The transmitter employs a free-running inductor–capacitor voltage-controlled oscillator (LC-VCO), a Gilbert mixer for OOK modulation and reverse isolation, and a current-reuse stacked power amplifier. It consumes 8 mA from a 3.3 V supply (26.4 mW), demonstrates modulation and receiver frame acquisition at a maximum raw input rate of 90 Mbps, corresponding to a 180 Mbps Manchester-coded line rate, and tunes from 3.266 to 3.445 GHz. End-to-end bit error rate (BER) was measured at raw rates of 15 and 31.2 Mbps, corresponding to encoded rates of 30 and 62.4 Mbps; the latter matches the in vivo data stream. The transmitter was integrated with a 128-channel recording chip and evaluated in freely moving adult C57 mice. Wireless hippocampal spike and local field potential (LFP) recordings, wired-system comparison, and event-locked LFP analysis support its feasibility for untethered neural recording. Full article

Empirical comparisons between cloud-based and on-premise database deployments under realistic e-commerce workloads remain limited. This study presents a controlled experimental evaluation of Microsoft SQL Server 2022 (on-premise) versus Azure SQL Database, using an identical .NET 8 e-commerce application (ASP.NET Core Web API, Blazor WebAssembly) with the same architecture, schema, and dataset (5000 product records, 10,000 transaction records). Performance was evaluated for SELECT, INSERT, UPDATE, and DELETE operations under workloads of up to 50 concurrent users, with each operation repeated 30 times, measuring query response time, throughput, and CPU utilization. Statistical analysis used repeated-measures ANOVA with Greenhouse–Geisser correction, Bonferroni-adjusted post hoc comparisons, and independent-samples t-tests (p < 0.001), with effect sizes reported using Cohen’s d. Azure SQL Database consistently outperformed the on-premise deployment: average SELECT response time decreased by 55% (203.5 ms vs. 452.3 ms), and throughput increased by 101.6% (987.6 vs. 489.8 operations/s). Although Azure showed higher average CPU utilization (20.4% vs. 4.9%), this reflects its dynamic resource allocation rather than reduced efficiency. Stress testing with 100,000 product records, 500,000 transaction records, and up to 500 concurrent users confirmed Azure’s superior scalability, reducing peak latency from 4850.7 ms to 580.4 ms. These findings provide strong empirical evidence supporting cloud migration for high-transaction e-commerce applications requiring low latency and high throughput. Full article