arXiv subjects

Shimeng Yu

Publications and source records attributed to Shimeng Yu.

At least 37 records · Page 2Linked to original sources

arXiv2025

Row Hammer Effect and Floating Body Effect of Monolithic 3D Stackable 1T1C DRAM

Monolithic 3D stackable 1T1C DRAM technology is on the rise, with initial prototypes reported by the industry. This work presents a comprehensive reliability study focusing on the intricate interplay between the row hammer effect and the floating body effect. First, using a TCAD model of a 3D DRAM mini-array, we categorize different cases of adjacent cells and show that the notorious row hammer effect induced by charge migration is significantly mitigated compared to 2D DRAM. However, we found that when incorporating an impact ionization model to account for the floating body characteristics of the silicon access transistor, the capacitive coupling between vertically stacked cells is severely exacerbated. Second, we conduct an in-depth investigation into the floating body effect itself. We systematically examine the dependence of this effect on key device parameters, including body thickness, doping concentration, and gate work function.

physics.app-ph↗

arXiv2025

Cryogenic Characterization of Ferroelectric Non-volatile Capacitors

Ferroelectric-based capacitive crossbar arrays have been proposed for energy-efficient in-memory computing in the charge domain. They combat the challenges like sneak paths and high static power faced by resistive crossbar arrays but are susceptible to thermal noise limiting the effective number of bits (ENOB) for the weighted sum. A direct way to reduce this thermal noise is by lowering the temperature as thermal noise is proportional to temperature. In this work, we first characterize the non-volatile capacitors (nvCaps) on a foundry 28 nm platform at cryogenic temperatures to evaluate the memory window, ON state retention as a function of temperature down to 77K, and then use the calibrated device models to simulate the capacitive crossbar arrays in SPICE at lower temperatures to demonstrate higher ENOB (~5 bits) for 128x128 multiple-and-accumulate (MAC) operations.

cs.ET↗

arXiv2025

FeNOMS: Enhancing Open Modification Spectral Library Search with In-Storage Processing on Ferroelectric NAND (FeNAND) Flash

The rapid expansion of mass spectrometry (MS) data, now exceeding hundreds of terabytes, poses significant challenges for efficient, large-scale library search - a critical component for drug discovery. Traditional processors struggle to handle this data volume efficiently, making in-storage computing (ISP) a promising alternative. This work introduces an ISP architecture leveraging a 3D Ferroelectric NAND (FeNAND) structure, providing significantly higher density, faster speeds, and lower voltage requirements compared to traditional NAND flash. Despite its superior density, the NAND structure has not been widely utilized in ISP applications due to limited throughput associated with row-by-row reads from serially connected cells. To overcome these limitations, we integrate hyperdimensional computing (HDC), a brain-inspired paradigm that enables highly parallel processing with simple operations and strong error tolerance. By combining HDC with the proposed dual-bound approximate matching (D-BAM) distance metric, tailored to the FeNAND structure, we parallelize vector computations to enable efficient MS spectral library search, achieving 43x speedup and 21x higher energy efficiency over state-of-the-art 3D NAND methods, while maintaining comparable accuracy.

cs.AR↗

arXiv2025

Stratum: System-Hardware Co-Design with Tiered Monolithic 3D-Stackable DRAM for Efficient MoE Serving

As Large Language Models (LLMs) continue to evolve, Mixture of Experts (MoE) architecture has emerged as a prevailing design for achieving state-of-the-art performance across a wide range of tasks. MoE models use sparse gating to activate only a handful of expert sub-networks per input, achieving billion-parameter capacity with inference costs akin to much smaller models. However, such models often pose challenges for hardware deployment due to the massive data volume introduced by the MoE layers. To address the challenges of serving MoE models, we propose Stratum, a system-hardware co-design approach that combines the novel memory technology Monolithic 3D-Stackable DRAM (Mono3D DRAM), near-memory processing (NMP), and GPU acceleration. The logic and Mono3D DRAM dies are connected through hybrid bonding, whereas the Mono3D DRAM stack and GPU are interconnected via silicon interposer. Mono3D DRAM offers higher internal bandwidth than HBM thanks to the dense vertical interconnect pitch enabled by its monolithic structure, which supports implementations of higher-performance near-memory processing. Furthermore, we tackle the latency differences introduced by aggressive vertical scaling of Mono3D DRAM along the z-dimension by constructing internal memory tiers and assigning data across layers based on access likelihood, guided by topic-based expert usage prediction to boost NMP throughput. The Stratum system achieves up to 8.29x improvement in decoding throughput and 7.66x better energy efficiency across various benchmarks compared to GPU baselines.

cs.AR↗

arXiv2025

Hardware Acceleration of Kolmogorov-Arnold Network (KAN) in Large-Scale Systems

Recent developments have introduced Kolmogorov-Arnold Networks (KAN), an innovative architectural paradigm capable of replicating conventional deep neural network (DNN) capabilities while utilizing significantly reduced parameter counts through the employment of parameterized B-spline functions with trainable coefficients. Nevertheless, the B-spline functional components inherent to KAN architectures introduce distinct hardware acceleration complexities. While B-spline function evaluation can be accomplished through look-up table (LUT) implementations that directly encode functional mappings, thus minimizing computational overhead, such approaches continue to demand considerable circuit infrastructure, including LUTs, multiplexers, decoders, and related components. This work presents an algorithm-hardware co-design approach for KAN acceleration. At the algorithmic level, techniques include Alignment-Symmetry and PowerGap KAN hardware aware quantization, KAN sparsity aware mapping strategy, and circuit-level techniques include N:1 Time Modulation Dynamic Voltage input generator with analog-compute-in-memory (ACIM) circuits. This work conducts evaluations on large-scale KAN networks to validate the proposed methodologies. Non-ideality factors, including partial sum deviations from process variations, have been evaluated with statistics measured from the TSMC 22nm RRAM-ACIM prototype chips. Utilizing optimally determined KAN hyperparameters in conjunction with circuit optimizations fabricated at the 22nm technology node, despite the parameter count for large-scale tasks in this work increasing by 500Kx to 807Kx compared to tiny-scale tasks in previous work, the area overhead increases by only 28Kx to 41Kx, with power consumption rising by merely 51x to 94x, while accuracy degradation remains minimal at 0.11% to 0.23%, demonstrating the scaling potential of our proposed architecture.

cs.AR↗

arXiv2025

Architecting Long-Context LLM Acceleration with Packing-Prefetch Scheduler and Ultra-Large Capacity On-Chip Memories

Long-context Large Language Model (LLM) inference faces increasing compute bottlenecks as attention calculations scale with context length, primarily due to the growing KV-cache transfer overhead that saturates High Bandwidth Memory (HBM). While prefetching techniques mitigate cache misses by fetching KV data in advance, their spatial and temporal benefits present new opportunities to exploit. This work proposes a packing-prefetch scheduling architecture with monolithic 3D (M3D) back-end-of-line (BEOL) compatible embedded memories with ultra-large on-chip capacity to accelerate long-context LLM inference. Our optimizations demonstrate 8.06x decode speedup and 1.83x overall latency reduction on Llama3.1-8B using TPUv6e-like hardware with additional 512MB BEOL memories over the serial execution. Evaluations of multi-request workloads on TPU-like architectures show 1.7x-2.4x throughput improvement and 1.5x-2.4x HBM bandwidth reduction compared to packing-only methods on Llama3.1-8B and Llama3.1-70B models. With the co-design of packing, prefetching, and BEOL memories, our approach alleviates HBM constraints and enables efficient long-context LLM inference.

cs.AR↗

arXiv2025

3DGauCIM: Accelerating Static/Dynamic 3D Gaussian Splatting via Digital CIM for High Frame Rate Real-Time Edge Rendering

Dynamic 3D Gaussian splatting (3DGS) extends static 3DGS to render dynamic scenes, enabling AR/VR applications with moving objects. However, implementing dynamic 3DGS on edge devices faces challenges: (1) Loading all Gaussian parameters from DRAM for frustum culling incurs high energy costs. (2) Increased parameters for dynamic scenes elevate sorting latency and energy consumption. (3) Limited on-chip buffer capacity with higher parameters reduces buffer reuse, causing frequent DRAM access. (4) Dynamic 3DGS operations are not readily compatible with digital compute-in-memory (DCIM). These challenges hinder real-time performance and power efficiency on edge devices, leading to reduced battery life or requiring bulky batteries. To tackle these challenges, we propose algorithm-hardware co-design techniques. At the algorithmic level, we introduce three optimizations: (1) DRAM-access reduction frustum culling to lower DRAM access overhead, (2) Adaptive tile grouping to enhance on-chip buffer reuse, and (3) Adaptive interval initialization Bucket-Bitonic sort to reduce sorting latency. At the hardware level, we present a DCIM-friendly computation flow that is evaluated using the measured data from a 16nm DCIM prototype chip. Our experimental results on Large-Scale Real-World Static/Dynamic Datasets demonstrate the ability to achieve high frame rate real-time rendering exceeding 200 frame per second (FPS) with minimal power consumption, merely 0.28 W for static Large-Scale Real-World scenes and 0.63 W for dynamic Large-Scale Real-World scenes. This work successfully addresses the significant challenges of implementing static/dynamic 3DGS technology on resource-constrained edge devices.

cs.AR↗

arXiv2025

A3D-MoE: Acceleration of Large Language Models with Mixture of Experts via 3D Heterogeneous Integration

Conventional large language models (LLMs) are equipped with dozens of GB to TB of model parameters, making inference highly energy-intensive and costly as all the weights need to be loaded to onboard processing elements during computation. Recently, the Mixture-of-Experts (MoE) architecture has emerged as an efficient alternative, promising efficient inference with less activated weights per token. Nevertheless, fine-grained MoE-based LLMs face several challenges: 1) Variable workloads during runtime create arbitrary GEMV-GEMM ratios that reduce hardware utilization, 2) Traditional MoE-based scheduling for LLM serving cannot fuse attention operations with MoE operations, leading to increased latency and decreased hardware utilization, and 3) Despite being more efficient than conventional LLMs, loading experts from DRAM still consumes significant energy and requires substantial DRAM bandwidth. Addressing these challenges, we propose: 1) A3D-MoE, a 3D Heterogeneous Integration system that employs state-of-the-art vertical integration technology to significantly enhance memory bandwidth while reducing Network-on-Chip (NoC) overhead and energy consumption. 2) A 3D-Adaptive GEMV-GEMM-ratio systolic array with V-Cache efficient data reuse and a novel unified 3D dataflow to solve the problem of reduced hardware utilization caused by arbitrary GEMV-GEMM ratios from different workloads, 3) A Hardware resource-aware operation fusion scheduler that fuses attention operations with MoE operations to enhance hardware performance, and 4) MoE Score-Aware HBM access reduction with even-odd expert placement that reduces DRAM access and bandwidth requirements. Our evaluation results indicate that A3D-MoE delivers significant performance enhancements, reducing latency by a factor of 1.8x to 2x and energy consumption by 2x to 4x, while improving throughput by 1.44x to 1.8x compared to the state-of-the-art.

cs.AR↗

arXiv2025

SLIM: A Heterogeneous Accelerator for Edge Inference of Sparse Large Language Model via Adaptive Thresholding

Large language models (LLMs) have demonstrated exceptional proficiency in understanding and generating human language, but efficient inference on resource-constrained embedded devices remains challenging due to large model sizes and memory-intensive operations in feedforward network (FFN) and multi-head attention (MHA) layers. While existing accelerators offload LLM inference to expensive heterogeneous computing systems, they fail to exploit the significant sparsity inherent in LLM operations, leaving hardware resources underutilized. We propose SLIM, an algorithm-hardware co-design optimized for sparse LLM serving on edge devices. SLIM exploits LLM sparsity through an adaptive thresholding algorithm that enables runtime-configurable sparsity with negligible accuracy loss, fetching only activated neurons to dramatically reduce data movement. Our heterogeneous hardware architecture strategically combines near-storage processing (NSP) and processing-in-memory (PIM): FFN weights are stored in high-density 3D NAND and computed using NSP units, while memory-intensive MHA operations are processed in PIM modules. This design significantly reduces memory footprint, data movement, and energy consumption. Our comprehensive evaluation demonstrates SLIM's effectiveness, achieving 13-18x throughput improvements over SSD-GPU systems and 9-10x better energy efficiency over DRAM-GPU systems while maintaining low latency, making cost-effective LLM deployment viable for edge computing environments.

cs.AR↗

arXiv2025

CMOS+X: Stacking Persistent Embedded Memories based on Oxide Transistors upon GPGPU Platforms

In contemporary general-purpose graphics processing units (GPGPUs), the continued increase in raw arithmetic throughput is constrained by the capabilities of the register file (single-cycle) and last-level cache (high bandwidth), which require the delivery of operands at a cadence demanded by wide single-instruction multiple-data (SIMD) lanes. Enhancing the capacity, density, or bandwidth of these memories can unlock substantial performance gains; however, the recent stagnation of SRAM bit-cell scaling leads to inequivalent losses in compute density. To address the challenges posed by SRAM's scaling and leakage power consumption, this paper explores the potential CMOS+X integration of amorphous oxide semiconductor (AOS) transistors in capacitive, persistent memory topologies (e.g., 1T1C eDRAM, 2T0C/3T0C Gain Cell) as alternative cells in multi-ported and high-bandwidth banked GPGPU memories. A detailed study of the density and energy tradeoffs of back-end-of-line (BEOL) integrated memories utilizing monolithic 3D (M3D)-integrated multiplexed arrays is conducted, while accounting for the macro-level limitations of integrating AOS candidate structures proposed by the device community (an aspect often overlooked in prior work). By exploiting the short lifetime of register operands, we propose a multi-ported AOS gain-cell capable of delivering 3x the read ports in ~76% of the footprint of SRAM with over 70% lower standby power, enabling enhancements to compute capacity, such as larger warp sizes or processor counts. Benchmarks run on a validated NVIDIA Ampere-class GPU model, using a modified version of Accel-Sim, demonstrate improvements of up to 5.2x the performance per watt and an average 8% higher geometric mean instruction per cycle (IPC) on various compute- and memory-bound tasks.

cs.ET↗

arXiv2025

An In-Situ Spatial-Temporal Sequence Detector for Neuromorphic Vision Sensor Empowered by High Density Vertical NAND Storage

Neuromorphic vision sensors require efficient real-time pattern recognition, yet conventional architectures struggle with energy and latency constraints. Here, we present a novel in-situ spatiotemporal sequence detector that leverages vertical NAND storage to achieve massively parallel pattern detection. By encoding each cell with two single-transistor-based multi-level cell (MLC) memory elements, such as ferroelectric field-effect transistors (FeFETs), and mapping a pixel's temporal sequence onto consecutive word lines (WLs), we enable direct temporal pattern detection within NAND strings. Each NAND string serves as a dedicated reference for a single pixel, while different blocks store patterns for distinct pixels, allowing large-scale spatial-temporal pattern recognition via simple direct bit-line (BL) sensing, a well-established operation in vertical NAND storage. We experimentally validate our approach at both the cell and array levels, demonstrating that vertical NAND-based detector achieves more than six orders of magnitude improvement in energy efficiency and more than three orders of magnitude reduction in latency compared to conventional CPU-based methods. These findings establish vertical NAND storage as a scalable and energy-efficient solution for next-generation neuromorphic vision processing.

cs.ET↗

arXiv2025

Optimization and Benchmarking of Monolithically Stackable Gain Cell Memory for Last-Level Cache

The Last Level Cache (LLC) is the processor's critical bridge between on-chip and off-chip memory levels - optimized for high density, high bandwidth, and low operation energy. To date, high-density (HD) SRAM has been the conventional device of choice; however, with the slowing of transistor scaling, as reflected in the industry's almost identical HD SRAM cell size from 5 nm to 3 nm, alternative solutions such as 3D stacking with advanced packaging like hybrid bonding are pursued (as demonstrated in AMD's V-cache). Escalating data demands necessitate ultra-large on-chip caches to decrease costly off-chip memory movement, pushing the exploration of device technology toward monolithic 3D (M3D) integration where transistors can be stacked in the back-end-of-line (BEOL) at the interconnect level. M3D integration requires fabrication techniques compatible with a low thermal budget (<400 degC). Among promising BEOL device candidates are amorphous oxide semiconductor (AOS) transistors, particularly desirable for their ultra-low leakage ( seconds) when used in a gain-cell configuration. This paper examines device, circuit, and system-level tradeoffs when optimizing BEOL-compatible AOS-based 2-transistor gain cell (2T-GC) for LLC. A cache early-exploration tool, NS-Cache, is developed to model caches in advanced 7 and 3 nm nodes and is integrated with the Gem5 simulator to systematically benchmark the impact of the newfound density/performance when compared to HD-SRAM, MRAM, and 1T1C eDRAM alternatives for LLC.

cs.ET↗

arXiv2025

Monolithic 3D FPGAs Utilizing Back-End-of-Line Configuration Memories

This work presents a novel monolithic 3D (M3D) FPGA architecture that leverages stackable back-end-of-line (BEOL) transistors to implement configuration memory and pass gates, significantly improving area, latency, and power efficiency. By integrating n-type (W-doped In_2O_3) and p-type (SnO) amorphous oxide semiconductor (AOS) transistors in the BEOL, Si SRAM configuration bits are substituted with a less leaky equivalent that can be programmed at logic-compatible voltages. BEOL-compatible AOS transistors are currently under extensive research and development in the device community, with investment by leading foundries, from which reported data is used to develop robust physics-based models in TCAD that enable circuit design. The use of AOS pass gates reduces the overhead of reconfigurable circuits by mapping FPGA switch block (SB) and connection block (CB) matrices above configurable logic blocks (CLBs), thereby increasing the proximity of logic elements and reducing latency. By interfacing with the latest Verilog-to-Routing (VTR) suite, an AOS-based M3D FPGA design implemented in 7 nm technology is demonstrated with 3.4x lower area-time squared product (AT^2), 27% lower critical path latency, and 26% lower reconfigurable routing block power on benchmarks including hyperdimensional computing and large language models (LLMs).

cs.ET↗

arXiv2024

Hardware Acceleration of Kolmogorov-Arnold Network (KAN) for Lightweight Edge Inference

Recently, a novel model named Kolmogorov-Arnold Networks (KAN) has been proposed with the potential to achieve the functionality of traditional deep neural networks (DNNs) using orders of magnitude fewer parameters by parameterized B-spline functions with trainable coefficients. However, the B-spline functions in KAN present new challenges for hardware acceleration. Evaluating the B-spline functions can be performed by using look-up tables (LUTs) to directly map the B-spline functions, thereby reducing computational resource requirements. However, this method still requires substantial circuit resources (LUTs, MUXs, decoders, etc.). For the first time, this paper employs an algorithm-hardware co-design methodology to accelerate KAN. The proposed algorithm-level techniques include Alignment-Symmetry and PowerGap KAN hardware aware quantization, KAN sparsity aware mapping strategy, and circuit-level techniques include N:1 Time Modulation Dynamic Voltage input generator with analog-CIM (ACIM) circuits. The impact of non-ideal effects, such as partial sum errors caused by the process variations, has been evaluated with the statistics measured from the TSMC 22nm RRAM-ACIM prototype chips. With the best searched hyperparameters of KAN and the optimized circuits implemented in 22 nm node, we can reduce hardware area by 41.78x, energy by 77.97x with 3.03% accuracy boost compared to the traditional DNN hardware.

cs.AR↗

arXiv2024

Towards Reverse-Engineering the Brain: Brain-Derived Neuromorphic Computing Approach with Photonic, Electronic, and Ionic Dynamicity in 3D integrated circuits

The human brain has immense learning capabilities at extreme energy efficiencies and scale that no artificial system has been able to match. For decades, reverse engineering the brain has been one of the top priorities of science and technology research. Despite numerous efforts, conventional electronics-based methods have failed to match the scalability, energy efficiency, and self-supervised learning capabilities of the human brain. On the other hand, very recent progress in the development of new generations of photonic and electronic memristive materials, device technologies, and 3D electronic-photonic integrated circuits (3D EPIC ) promise to realize new brain-derived neuromorphic systems with comparable connectivity, density, energy-efficiency, and scalability. When combined with bio-realistic learning algorithms and architectures, it may be possible to realize an 'artificial brain' prototype with general self-learning capabilities. This paper argues the possibility of reverse-engineering the brain through architecting a prototype of a brain-derived neuromorphic computing system consisting of artificial electronic, ionic, photonic materials, devices, and circuits with dynamicity resembling the bio-plausible molecular, neuro/synaptic, neuro-circuit, and multi-structural hierarchical macro-circuits of the brain based on well-tested computational models. We further argue the importance of bio-plausible local learning algorithms applicable to the neuromorphic computing system that capture the flexible and adaptive unsupervised and self-supervised learning mechanisms central to human intelligence. Most importantly, we emphasize that the unique capabilities in brain-derived neuromorphic computing prototype systems will enable us to understand links between specific neuronal and network-level properties with system-level functioning and behavior.

cs.ET↗

arXiv2024

Paving the Way for Pass Disturb Free Vertical NAND Storage via A Dedicated and String-Compatible Pass Gate

In this work, we propose a dual-port cell design to address the pass disturb in vertical NAND storage, which can pass signals through a dedicated and string-compatible pass gate. We demonstrate that: i) the pass disturb-free feature originates from weakening of the depolarization field by the pass bias at the high-${V}_{TH}$ (HVT) state and the screening of the applied field by channel at the low-${V}_{TH}$ (LVT) state; ii) combined simulations and experimental demonstrations of dual-port design verify the disturb-free operation in a NAND string, overcoming a key challenge in single-port designs; iii) the proposed design can be incorporated in a highly scaled vertical NAND FeFET string and the pass gate can be incorporated into the existing 3D NAND with the negligible overhead of the pass gate interconnection through a global bottom pass gate contact in the substrate.

cs.ET↗

arXiv2022

A wearable sensor vest for social humanoid robots with GPGPU, IoT, and modular software architecture

Currently, most social robots interact with their surroundings and humans through sensors that are integral parts of the robots, which limits the usability of the sensors, human-robot interaction, and interchangeability. A wearable sensor garment that fits many robots is needed in many applications. This article presents an affordable wearable sensor vest, and an open-source software architecture with the Internet of Things (IoT) for social humanoid robots. The vest consists of touch, temperature, gesture, distance, vision sensors, and a wireless communication module. The IoT feature allows the robot to interact with humans locally and over the Internet. The designed architecture works for any social robot that has a general-purpose graphics processing unit (GPGPU), I2C/SPI buses, Internet connection, and the Robotics Operating System (ROS). The modular design of this architecture enables developers to easily add/remove/update complex behaviors. The proposed software architecture provides IoT technology, GPGPU nodes, I2C and SPI bus mangers, audio-visual interaction nodes (speech to text, text to speech, and image understanding), and isolation between behavior nodes and other nodes. The proposed IoT solution consists of related nodes in the robot, a RESTful web service, and user interfaces. We used the HTTP protocol as a means of two-way communication with the social robot over the Internet. Developers can easily edit or add nodes in C, C++, and Python programming languages. Our architecture can be used for designing more sophisticated behaviors for social humanoid robots.

cs.RO↗

arXiv2021

Logic Compatible High-Performance Ferroelectric Transistor Memory

Silicon ferroelectric field-effect transistors (FeFETs) with low-k interfacial layer (IL) between ferroelectric gate stack and silicon channel suffers from high write voltage, limited write endurance and large read-after-write latency due to early IL breakdown and charge trapping and detrapping at the interface. We demonstrate low voltage, high speed memory operation with high write endurance using an IL-free back-end-of-line (BEOL) compatible FeFET. We fabricate IL-free FeFETs with 28nm channel length and 126nm width under a thermal budget <400C by integrating 5nm thick Hf0.5Zr0.5O2 gate stack with amorphous Indium Tungsten Oxide (IWO) semiconductor channel. We report 1.2V memory window and read current window of 10^5 for program and erase, write latency of 20ns with +/-2V write pulses, read-after-write latency <200ns, write endurance cycles exceeding 5x10^10 and 2-bit/cell programming capability. Array-level analysis establishes IL-free BEOL FeFET as a promising candidate for logic-compatible high-performance on-chip buffer memory and multi-bit weight cell for compute-in-memory accelerators.

cond-mat.mes-hall↗