Fast Memory Research Articles

Origin of charges in bulk Si:HfO2 FeFET probed by nanosecond polarization measurements

FeFET technology offers the potential for fast, energy-efficient, low-cost, and high-capacity non-volatile memory and neuromorphic devices. However, charge trapping significantly affects device operation, leading to issues like read-after-write delay and limited endurance. Therefore, a detailed understanding of charge trapping, charge origin and its role in polarization switching is crucial. In this study, we uncover the spectral energy origin of polarization charges in Si:HfO2 N-FeFET by probing electron (conduction band) and hole (valence band) currents separately during polarization-voltage (P–V) measurements. We utilize a fast (∼20 ns) and modified positive-up-negative-down (PUND) technique, where bulk, source, and drain currents of the FeFET are measured separately. The nanosecond timescale of the measurement results in measurable currents in FeFETs having dimensions of a few μm. This charge separation shows that program (PRG, VGS > 0) charge originates from the conduction band, whereas erase (ERS, VGS < 0) originates from the valence band of the Si. Moreover, the polarization curve (P–V) of a cycled device (following 5000 PRG/ERS pulses) shows measurable hysteresis even though the transfer curve of the same device shows that the memory window in the threshold voltage vanishes. Therefore, the FeFET polarization state can be read without delay after write operation by the fast PUND measurement, both for pristine and cycled FeFETs.

4D Printing of Multifunctional and Biodegradable PLA-PBAT-Fe3O4 Nanocomposites with Supreme Mechanical and Shape Memory Properties.

4D printing magneto-responsive shape memory polymers (SMPs) using biodegradable nanocomposites can overcome their low toughness and thermal resistance, and produce smart materials that can be controlled remotely without contact. This study presented the development of 3D/4D printable nanocomposites based on poly (lactic acid) (PLA)-poly (butylene adipate-co-terephthalate) (PBAT) blends and magnetite (Fe3O4) nanoparticles. The nanocomposites are prepared by melt mixing PLA-PBAT blends with different Fe3O4 contents (10, 15, and 20 wt%) and extruded into granules for material extrusion 3D printing. The morphology, dynamic mechanical thermal analysis (DMTA), mechanical properties, and shape memory behavior of the nanocomposites are investigated. The results indicated that the Fe3O4 nanoparticles are preferentially distributed in the PBAT phases, enhancing the storage modulus, thermal stability, strength, elongation, toughness, shape fixity, and recovery of the nanocomposites. The optimal Fe3O4 loading is found to be 10 wt%, as higher loadings led to nanoparticle agglomeration and reduced performance. The nanocomposites also exhibited fast shape memory response under thermal and magnetic activation due to the presence of Fe3O4 nanoparticles. The 3D/4D printable nanocomposites demonstrated multifunctional multi-trigger shape-memory capabilities and potential applications in contactless and safe actuation.

Adaptive Quotient Filters

Filters trade off accuracy for space and occasionally return false positive matches with a bounded error. Numerous systems use filters in fast memory to avoid performing expensive I/Os to slow storage. A fundamental limitation in traditional filters is that they do not change their representation upon seeing a false positive match. Therefore, the maximum false positive rate is only guaranteed for a single query, not for an arbitrary set of queries. We can improve the filter's performance on a stream of queries, especially on a skewed distribution, if we can adapt after encountering false positives. Adaptive filters, such as telescoping quotient filters and adaptive cuckoo filters, update their representation upon detecting a false positive to avoid repeating the same error in the future. Adaptive filters require an auxiliary structure, typically much larger than the main filter and often residing on slow storage, to facilitate adaptation. However, existing adaptive filters are not practical and have not been adopted in real-world systems for two main reasons. First, they offer weak adaptivity guarantees, meaning that fixing a new false positive can cause a previously fixed false positive to come back. Secondly, the sub-optimal design of the auxiliary structure results in adaptivity overheads so substantial that they can actually diminish overall system performance compared to a traditional filter. In this paper, we design and implement the \sysname, the first practical adaptive filter with minimal adaptivity overhead and strong adaptivity guarantees, which means that the performance and false-positive guarantees continue to hold even for adversarial workloads. The \sysname is based on the state-of-the-art quotient filter design and preserves all the critical features of the quotient filter such as cache efficiency and mergeability. Furthermore, we employ a new auxiliary structure design which results in considerably low adaptivity overhead and makes the \sysname practical in real systems. We evaluate the \sysname by using it to filter queries to an on-disk B-tree database and find no negative impact on insert or query performance compared to traditional filters. Against adversarial workloads, the \sysname preserves system performance, whereas traditional filters incur 2× slowdown from adversaries representing as low as 1% of the workload. Finally, we show that on skewed query workloads, the \sysname can reduce the false-positive rate 100× using negligible (1/1000th of a bit per item) space overhead.

Shunt-free cryogenic memory using ferromagnetic insulator-based Josephson junctions

Magnetic Josephson junctions are the preferred candidate devices for designing fast and scalable cryogenic memory elements. This is especially the case for rapid single-flux quantum-based superconducting electronics, where the speed mismatch between logic and memory elements have remained a long-standing challenge. In this Letter, we demonstrate a simple tri-layer Josephson memory device using ferromagnetic insulating (FI) GdN-based S/FI/S vertical mesa-type junctions, with reliable nonvolatile memory operation without the need of a shunt resistor at 4.2 K. The characteristic frequency of our devices is approximately 90 GHz, corresponding to an IcRn product of 177 μV. We demonstrate a thorough study of the parameter spaces required for designing these devices and identify the scope for future improvements that can lead to further miniaturization and higher operating speed of these devices.

Quantized CNN-based efficient hardware architecture for real-time hand gesture recognition

Nowadays, Convolutional Neural Networks (CNN) have been widely adopted for vision-based hand gesture recognition. Several existing CNN architectures designed for gesture classification perform well with high accuracy but require a high memory footprint and processing when deployed on low-power embedded devices. To address this issue, we present a quantized CNN-based efficient framework for meeting real-life hand gesture recognition challenges. The proposed quantized CNN is designed and implemented using the FINN-based pipelined streaming architecture on an FPGA. Moreover, hardware-based optimizations are used to minimize the resources needed and to achieve fast memory access. Experimental results demonstrate that the developed recognition system achieves an average accuracy of 92% on the numeral database of Indian Sign Language (ISL). Additionally, our optimized design attains an inference latency of 0.85ms for real-time single gesture prediction on the PYNQ Zynq Ultrascale (ZU) FPGA, consuming only 3.63 W of power. The proposed design achieves a better trade-off between hardware resource utilization and speed performance, over previous designs.

Association Between Subclinical Thyroid Dysfunction and Cognitive Decline: Findings From the ELSA-Brasil Study.

Thyroid dysfunction has been associated with cognitive decline and dementia. However, the role of subtle thyroid hormone alterations in cognitive function is still debatable. Participants without overt thyroid dysfunction aged 35-74 years at baseline were evaluated in 3 study waves (2008-2010, 2012-2014, and 2017-2019). We assessed baseline thyroid-stimulating hormone (TSH), free thyroxine (FT4), and free triiodothyronine (FT3). Cognitive performance was evaluated every 4 years in each wave using 10-word immediate and late recall, word recognition, semantic (animals category) and phonemic (letter f) verbal fluency, and the trail-making B-version tests. A global composite z-score was derived from these tests. The associations of TSH, FT4, and FT3 levels with cognitive decline over time were evaluated using linear mixed-effect models adjusted for sociodemographic, clinical, and lifestyle variables. In 9524 participants (mean age 51.2 ± 8.9 years old, 51% women, 52% White), there was no association between baseline TSH, FT4, and FT3 levels and cognitive decline during the follow-up. However, increase in FT4 levels over time was associated with faster memory (β = -0.004, 95% CI = -0.007; -0.001, p = .014), verbal fluency (β = -0.003, 95% CI = -0.007; -0.0005, p = .021), executive function (β = -0.004, 95% CI = -0.011; -0.003, p < .001), and global cognition decline (β = -0.003, 95% CI = -0.006; -0.001, p = .001). Decrease in FT4 levels over time was associated with faster verbal fluency (β = -0.003, 95% CI = -0.007; -0.0004, p = .025) and executive function (β = -0.004, 95% CI = -0.007; -0.0003, p = .031) decline. An increase or decrease in FT4 levels over time was associated with faster cognitive decline in middle-aged and older adults without overt thyroid dysfunction during 8 years of follow-up.

PROMPT: A Fast and Extensible Memory Profiling Framework

Memory profiling captures programs’ dynamic memory behavior, assisting programmers in debugging, tuning, and enabling advanced compiler optimizations like speculation-based automatic parallelization. As each use case demands its unique program trace summary, various memory profiler types have been developed. Yet, designing practical memory profilers often requires extensive compiler expertise, adeptness in program optimization, and significant implementation effort. This often results in a void where aspirations for fast and robust profilers remain unfulfilled. To bridge this gap, this paper presents PROMPT, a framework for streamlined development of fast memory profilers. With PROMPT, developers need only specify profiling events and define the core profiling logic, bypassing the complexities of custom instrumentation and intricate memory profiling components and optimizations. Two state-of-the-art memory profilers were ported with PROMPT where all features preserved. By focusing on the core profiling logic, the code was reduced by more than 65% and the profiling overhead was improved by 5.3× and 7.1× respectively. To further underscore PROMPT’s impact, a tailored memory profiling workflow was constructed for a sophisticated compiler optimization client. In 570 lines of code, this redesigned workflow satisfies the client’s memory profiling needs while achieving more than 90% reduction in profiling overhead and improved robustness compared to the original profilers.

SEGUL: Ultrafast, memory-efficient and mobile-friendly software for manipulating and summarizing phylogenomic datasets.

Phylogenetic studies now routinely require manipulating and summarizing thousands of data files. For most of these tasks, currently available software requires considerable computing resources and substantial knowledge of command-line applications. We develop an ultrafast and memory-efficient software, SEGUL, that performs common phylogenomic dataset manipulations and calculates statistics summarizing essential data features. Our software is available as standalone command-line interface (CLI) and graphical user interface (GUI) applications, and as a library for Rust, R and Python, with possible support of other languages. The CLI and library versions run native on Windows, Linux and macOS, including Apple ARM Macs. The GUI version extends support to include mobile iOS, iPadOS and Android operating systems. SEGUL leverages the high performance of the Rust programming language to offer fast execution times and low memory footprints regardless of dataset size and platform choice. The inclusion of a GUI minimizes bioinformatics barriers to phylogenomics while SEGUL's efficiency reduces economic barriers by allowing analysis on inexpensive hardware. Our support for mobile operating systems further enables teaching phylogenomics where access to computing power is limited.

Selective Focus: Investigating Semantics Sensitivity in Post-training Quantization for Lane Detection

Lane detection (LD) plays a crucial role in enhancing the L2+ capabilities of autonomous driving, capturing widespread attention. The Post-Processing Quantization (PTQ) could facilitate the practical application of LD models, enabling fast speeds and limited memories without labeled data. However, prior PTQ methods do not consider the complex LD outputs that contain physical semantics, such as offsets, locations, etc., and thus cannot be directly applied to LD models. In this paper, we pioneeringly investigate semantic sensitivity to post-processing for lane detection with a novel Lane Distortion Score. Moreover, we identify two main factors impacting the LD performance after quantization, namely intra-head sensitivity and inter-head sensitivity, where a small quantization error in specific semantics can cause significant lane distortion. Thus, we propose a Selective Focus framework deployed with Semantic Guided Focus and Sensitivity Aware Selection modules, to incorporate post-processing information into PTQ reconstruction. Based on the observed intra-head sensitivity, Semantic Guided Focus is introduced to prioritize foreground-related semantics using a practical proxy. For inter-head sensitivity, we present Sensitivity Aware Selection, efficiently recognizing influential prediction heads and refining the optimization objectives at runtime. Extensive experiments have been done on a wide variety of models including keypoint-, anchor-, curve-, and segmentation-based ones. Our method produces quantized models in minutes on a single GPU and can achieve 6.4\% F1 Score improvement on the CULane dataset. Code and supplementary statement can be found at https://github.com/PannenetsF/SelectiveFocus.

SasWOT: Real-Time Semantic Segmentation Architecture Search WithOut Training

In this paper, we present SasWOT, the first training-free Semantic segmentation Architecture Search (SAS) framework via an auto-discovery proxy. Semantic segmentation is widely used in many real-time applications. For fast inference and memory efficiency, Previous SAS seeks the optimal segmenter by differentiable or RL Search. However, the significant computational costs of these training-based SAS limit their practical usage. To improve the search efficiency, we explore the training-free route but empirically observe that the existing zero-cost proxies designed on the classification task are sub-optimal on the segmentation benchmark. To address this challenge, we develop a customized proxy search framework for SAS tasks to augment its predictive capabilities. Specifically, we design the proxy search space based on the some observations: (1) different inputs of segmenter statistics can be well combined; (2) some basic operators can effectively improve the correlation. Thus, we build computational graphs with multiple statistics as inputs and different advanced basis arithmetic as the primary operations to represent candidate proxies. Then, we employ an evolutionary algorithm to crossover and mutate the superior candidates in the population based on correlation evaluation. Finally, based on the searched proxy, we perform the segmenter search without candidate training. In this way, SasWOT not only enables automated proxy optimization for SAS tasks but also achieves significant search acceleration before the retrain stage. Extensive experiments on Cityscapes and CamVid datasets demonstrate that SasWOT achieves superior trade-off between accuracy and speed over several state-of-the-art techniques. More remarkably, on Cityscapes dataset, SasWOT achieves the performance of 71.3% mIoU with the speed of 162 FPS.

The application of C/Sb composite multilayer films on fast flexible phase change memory

In this paper, the phase transformation properties of C/Sb composite multilayer films based on flexible substrates were studied. After different times of bending, the flexible films can still achieve amorphous to crystalline phase transition, but the crystalline resistance increases to different degrees. When the bending times are less than 15000 times, the deformation makes the grain fine, the surface of the film tends to be flat, the surface electric field effect is weakened, and the resistance drift is reduced. When the bending times are higher than 15000 times, cracks begin to appear in the film, and the cracks’ depth and width increase with the bending times. The distribution of C and Sb elements at the crack was analyzed by scanning electronic microscope energy spectrum. The phase change memory device of C/Sb composite multilayer film based on flexible PEEK substrate has been made, and the rapid reversible resistance transition (∼10 ns) in various complex states has been successfully achieved. This research provides a new idea for flexible information storage.

Hardware-Software Collaborative Tiered-Memory Management Framework for Virtualization

The tiered-memory system can effectively expand the memory capacity for virtual machines (VMs). However, virtualization introduces new challenges specifically in enforcing performance isolation, minimizing context switching, and providing resource overcommit. None of the state-of-the-art designs consider virtualization and address these challenges; we observe that a VM with tiered memory incurs up to a 2× slowdown compared to a DRAM-only VM. We propose vTMM , a hardware-software collaborative tiered-memory management framework for virtualization. A key insight in vTMM is to leverage the unique system features in virtualization to meet the above challenges. vTMM automatically determines page hotness and migrates pages between fast and slow memory to achieve better performance. Specially, vTMM optimizes page tracking and migration based on page-modification logging (PML), a hardware-assisted virtualization mechanism, and adaptively distinguishes hot/cold pages through the page “temperature” sorting. vTMM also dynamically adjusts fast memory among multi-VMs on demand by using a memory pool. Further, vTMM tracks huge pages at regular-page granularity in hardware and splits/merges pages in software, realizing hybrid-grained page management and optimization. We implement and evaluate vTMM with single-grained page management on an Intel processor, and the hybrid-grained page management on a Sunway processor with hardware mode supporting hardware/software co-designs. Experiments show that vTMM outperforms existing tiered-memory management designs in virtualization.

Toward Energy‐Efficient Ferroelectric Field‐Effect Transistors and Ferroelectric Random Access Memories: Tailoring the Coercive Field of Ferroelectric HfO2 Films

Many modern applications require fast and reliable nonvolatile memory. Ferroelectric (FE) memories, such as FE field‐effect transistors and FE random access memories, show great promise in meeting these requirements. Another key factor is power efficiency. In the case of FE memories, the coercive field is a critical parameter for improving power consumption. This article introduces and reviews four promising methods to tailor the of FE films based on Zr‐doped : 1) composites with various Hf contents, 2) hybrid stacks consisting of antiferroelectric sublayers with excess Zr (x 0.5) and FE sublayers, 3) superlattice structures (also known as nanolaminates) consisting of and sublayers stacked on top of each other, and 4) films doped with an aluminum impurities. Electrical characterization of metal–FE–metal capacitor test structures confirms the suitability of all approaches. Analysis of the strengths and weaknesses is conducted, considering the tuning range, shape stability of the polarization versus electric field hysteresis loop, and the impact of these approaches on the remanent polarization .

Microwave Field-Induced Changes in Raman Modes and Magnetic Force Images of Antiferromagnetic NiO Films

Effective control of domain walls or magnetic textures in antiferromagnets promises to enable robust, fast, and nonvolatile memories. The lack of net magnetic moment in antiferromagnets implies the need for creative ways to achieve such a manipulation. We conducted a study to investigate changes in magnetic force microscopy (MFM) imaging and in the magnon-related mode in Raman spectroscopy of virgin NiO films under a microwave pump. After MFM and Raman studies were conducted, a combined action of broadband microwave (0.01–20 GHz, power scanned from −20 to 5 dBm) and magnetic field (up to 3 kOe) were applied to virgin epitaxial (111) NiO and (100) NiO films grown on (0001) Al2O3 and (100) MgO substrates, following which the MFM and Raman studies were repeated. We observed a suppression of the magnon-related Raman mode subsequent to the microwave exposure. Based on MFM imaging, this effect appeared to be caused by the suppression of large antiferromagnetic domain walls due to the possible excitation of antiferromagnetic spin oscillations localized within the antiferromagnetic domain walls.

RmdnCache: Dual-Space Prefetching Neural Network for Large-Scale Volume Visualization.

Volume visualization plays a significant role in revealing important intrinsic patterns of 3D scientific datasets. However, these datasets are often large, making it challenging for interactive visualization systems to deliver a seamless user experience because of high input latency that arises from I/O bottlenecks and limited fast memory resources with high miss rates. To address this issue, we have proposed a deep learning-based prefetching method called RmdnCache, which optimizes the data flow across the memory hierarchy to reduce the input latency of large-scale volume visualization. Our approach accurately prefetches the content of the next view to fast memory using learning-based prediction while rendering the current view. The proposed deep learning architecture consists of two networks, RNN and MDN in respective spaces, which work together to predict both the location and likelihood distribution of the next view for defining an optimal prefetching range. Our method outperforms existing state-of-the-art prefetching algorithms in reducing overall input latency for visualizing real-world large-scale volumetric datasets.

Density functional theory study of the electronic and optical properties of pentagraphyne nanotubes.

Pentagraphyne (PG-yne), a recently predicted two-dimensional (2D) carbon allotrope with appealing properties, has opened up possibilities for a wide range of applications. In this study, we investigate the structural, electronic, optical, and electrical transport properties of a novel one-dimensional (1D) system called pentagraphyne nanotubes (PG-yneNTs), formed by rolling a PG-yne sheet, using density functional theory (DFT) calculations. We design PG-yneNTs with diameters ranging from 6.94 Å to 13.62 Å and employ state-of-the-art theoretical calculations to confirm their energetic, dynamic, and thermodynamic stability. Our electronic band structure calculations reveal that all these nanotubes are wide indirect band gap semiconductors. Remarkably, PG-yneNTs exhibit superior optical properties, including high absorption coefficients and absorption spectra covering the visible regime of the electromagnetic spectrum, making them potential candidates for visible-light-driven photocatalysis and solar cells. Interestingly, both the electronic and optical band gaps increase with the diameter of the nanotubes. Additionally, the observation of negative differential resistance (NDR) phenomena in (4, 0) PG-yneNT suggests their potential applications in NDR devices such as fast switches, frequency multipliers, and memory devices.

FPGA-Based Implementation of SCADA System for Fuel Management

With the emergence of Digital Processors and Logic Design technologies, like FPGAs, the trend of using IC-based plug-and-play modules has been minimized. The focus is commercial off-shelf designing and developing modules and sub-modules on new FPGA technology. Some of the critical factors that make FPGA more suitable include enormous computational power, capabilities to perform logic operations, high-speed clocks, fast memory, and various built-in primitives for computation-intensive arithmetic operations. FPGA becomes suitable for implementing data capturing and processing systems with these characteristics. This paper is about designing and developing an FPGA-based Supervisory Control and Data Acquisition (SCADA) System to monitor and control the fuel level transition between two tanks that generally are designed using PLC. PLC-based SCADA systems are easy to implement, but when discussing performance, PLC cannot replace FPGA. The FPGA chosen is Spartan-3 due to its low cost and meeting the industrial temperature needs. At the same time, it may be translated into any advanced FPGA, producing fewer sources and power consumption benefits. The simulator is Xilinx 14.7 ISE, along with SDK. The implemented system provides the inherent amenities of SCADA along with the benefits of high speed, more accuracy, reduced and predictable delay, and a purely digitized system facilitated by the FPGA.

A new strategy for reconfigurable actuators based on oxidation of fast responsive shape memory polyarylene sulfide sulfone

Actuators have shown great potential application in many fields. Till now, it remains a challenge to design and exploit diverse modes actuators which can endure in hard environment by reconfiguration. Here, a series of shape memory polyarylene sulfide sulfone (SMP-PASS) shows excellent thermal properties, good mechanical properties and good shape memory behaviors were synthesized. The shape recovery temperature of can further increased by oxidizing thioether bonds to sulfone groups by facile oxidation treatment. According to the different temperature stimulus response of the samples before and after oxidation, herein, we propose a simple strategy to fabricate reconfigurable actuators that can be used in extreme environments by patterns in both plane and depth directions of SMP-PASS films by post-oxidation modification. SMP-PASS films after programming anisotropic can transform planar films into various 3D structure actuators under heating stimulation, unambiguous confirming that the facile oxidation method supplies an efficient route to yield multiple and thermal-resistant actuators. That not only supplies a novel, re-processable and fast-responsive shape memory polymer candidate, but also provides a facile and effective strategy to fabricate smart 3D actuators.

Mechanistic description of spontaneous loss of memory persistent activity based on neuronal synaptic strength

Persistent neural activity associated with working memory (WM) lasts for a limited time duration. Current theories suggest that its termination is actively obtained via inhibitory currents, and there is currently no theory regarding the possibility of a passive memory-loss mechanism that terminates memory persistent activity. Here, we develop an analytical-framework, based on synaptic strength, and show via simulations and fitting to wet-lab experiments, that passive memory-loss might be a result of an ionic-current long-term plateau, i.e., very slow reduction of memory followed by abrupt loss. We describe analytically the plateau, when the memory state is just below criticality. These results, including the plateau, are supported by experiments performed on rats. Moreover, we show that even just above criticality, forgetfulness can occur due to neuronal noise with ionic-current fluctuations, yielding a plateau, representing memory with very slow decay, and eventually a fast memory decay. Our results could have implications for developing new medications, targeted against memory impairments, through modifying neuronal noise.

Critical Data Backup with Hybrid Flash-Based Consumer Devices

Hybrid flash-based storage constructed with high-density and low-cost flash memory has become increasingly popular in consumer devices in the last decade due to its low cost. However, its poor reliability is one of the major concerns. To protect critical data for guaranteeing user experience, some methods are proposed to improve the reliability of consumer devices with non-hybrid flash storage. However, with the widespread use of hybrid storage, these methods will result in severe problems, including significant performance and endurance degradation. This is caused by the fact that the different characteristics of flash memory in hybrid storage are not considered, e.g., performance, endurance, and access granularity. To address these problems, a critical data backup (CDB) design is proposed to ensure critical data reliability at a low cost. The basic idea is to accumulate two copies of critical data in the fast memory first to make full use of its performance and endurance. Then, one copy will be migrated to the slow memory in the stripe to avoid the write amplification caused by different access granularity between them. By respecting the different characteristics of flash memory in hybrid storage, CDB can achieve encouraging performance and endurance improvement compared with the state-of-the-art. Furthermore, to avoid performance and lifetime degradation caused by the backup data occupying too much space of fast memory, CDB Pro is designed. Two advanced schemes are integrated. One is making use of the pseudo-single-level-cell (pSLC) technique to make a part of slow memory become high-performance. By supplying some high-performance space, data will be fully updated before being evicted to slow memory. More invalid data are generated which reduces eviction costs. Another is to categorize data into three types according to their different life cycles. By putting the same type of data in a block, the eviction efficiency is improved. Therefore, both can improve device performance and lifetime based on CDB. Experiments are conducted to prove the efficiency of CDB and CDB Pro. Experimental results show that compared with the state-of-the-arts, CDB can ensure critical data reliability with lower device performance and lifetime loss whereas CDB Pro can diminish the loss further.