SearcharxivSearch

arXiv subjects

Sheldon X. -D. Tan

Publications and source records attributed to Sheldon X. -D. Tan.

13 recordsLinked to original sources

WarPGNN: A Parametric Thermal Warpage Analysis Framework with Physics-aware Graph Neural Network

With the advent of system-in-package (SiP) chiplet-based design and heterogeneous 2.5D/3D integration, thermal-induced warpage has become a critical reliability concern. While conventional numerical approaches can deliver highly accurate results, they often incur prohibitively high computational costs, limiting their scalability for complex chiplet-package systems. In this paper, we present WarPGNN, an efficient and accurate parametric thermal warpage analysis framework powered by Graph Neural Networks (GNNs). By operating directly on graphs constructed from the floorplans, WarPGNN enables fast warpage-aware floorplan exploration and exhibits strong transferability across diverse package configurations. Our method first encodes multi-die floorplans into reduced Transitive Closure Graphs (rTCGs), then a Graph Convolution Network (GCN)-based encoder extracts hierarchical structural features, followed by a U-Net inspired decoder that reconstructs warpage maps from graph feature embeddings. Furthermore, to address the long-tailed pattern of warpage data distribution, we developed a physics-informed loss and revised a message-passing encoder based on Graph Isomorphic Network (GIN) that further enhance learning performance for extreme cases and expressiveness of graph embeddings. Numerical results show that WarPGNN achieves more than 205.91x speedup compared with the 2-D efficient FEM-based method and over 119766.64x acceleration with 3-D FEM method COMSOL, respectively, while maintaining comparable accuracy at only 1.26% full-scale normalized RMSE and 2.21% warpage value error. Compared with recent DeepONet-based model, our method achieved comparable prediction accuracy and inference speedup with 3.4x lower training time. In addition, WarPGNN demonstrates remarkable transferability on unseen datasets with up to 3.69% normalized RMSE and similar runtime.

cs.AR

EMSpice 3: Full-chip Temperature-Aware Multiphysics Electromigration and IR-Drop Analysis

This paper presents EMSpice~3, a full-chip multiphysics framework for coupled electromigration (EM), thermomigration (TM), and IR-drop analysis of practical power-grid (P/G) networks. The framework is, to our knowledge, the first EM-IR analysis flow that jointly incorporates Joule heating and practical spatial thermal profiles for full-chip P/G network designs. It operates on extracted power-grid netlists and combines an immortality check, transient EM/TM stress evolution, void-induced resistance updates, repeated IR-drop recomputation, and optional Monte Carlo lifetime prediction. To make chip-level EM analysis tractable, the framework integrates an extended rational Krylov subspace method into the transient solver, achieving $1.18\times$--$1.50\times$ speedup with sub-0.05% reported TTF/final-IR metric error relative to the default non-Krylov FDTD analysis across six benchmark designs. The numerical results reveal that the specific spatial temperature profile can have a more significant impact on P/G network lifetime than the average temperature itself. In the RISC-V core, a higher-average-temperature profile can avoid the 10% IR-drop failure threshold when its hotspots are less aligned with critical current paths, while mapped temperature gradients can move the critical void location and change which resistor branches are degraded. Monte Carlo analysis further shows design-specific variation sensitivity: under 20% variation in EM diffusivity and critical stress, the RISC-V core exhibits about 15.8% TTF coefficient of variation, whereas the ARM Cortex-A logic core exhibits only 0.0058\%. These results show that practical thermal profiles, resistance feedback, and stochastic material variation must be considered jointly for predictive full-chip EM-IR analysis.

cs.AR

Accelerating Physics-Based Electromigration Analysis via Rational Krylov Subspaces

Electromigration (EM) induced stress evolution is a major reliability challenge in nanometer-scale VLSI interconnects. Accurate EM analysis requires solving stress-governing partial differential equations over large interconnect trees, which is computationally expensive using conventional finite-difference methods. This work proposes two fast EM stress analysis techniques based on rational Krylov subspace reduction. Unlike traditional Krylov methods that expand around zero frequency, rational Krylov methods enable expansion at selected time constants, aligning directly with metrics such as nucleation and steady-state times and producing compact reduced models with minimal accuracy loss. Two complementary frameworks are developed: a frequency-domain extended rational Krylov method, ExtRaKrylovEM, and a time-domain rational Krylov exponential integration method, EiRaKrylovEM. We show that the accuracy of both methods depends strongly on the choice of expansion point, or shift time, and demonstrate that effective shift times are typically close to times of interest such as nucleation or post-void steady state. Based on this observation, a coordinate descent optimization strategy is introduced to automatically determine optimal reduction orders and shift times for both nucleation and post-void phases. Experimental results on synthesized structures and industry-scale power grids show that the proposed methods achieve orders-of-magnitude improvements in efficiency and accuracy over finite-difference solutions. Using only 4 to 6 Krylov orders, the methods achieve sub-0.1 percent error in nucleation time and resistance change predictions while delivering 20 to 500 times speedup. In contrast, standard extended Krylov methods require more than 50 orders and still incur 10 to 20 percent nucleation time error, limiting their practicality for EM-aware optimization and stochastic EM analysis.

cs.AR

BPINN-EM-Post: Bayesian Physics-Informed Neural Network based Stochastic Electromigration Damage Analysis in the Post-void Phase

In contrast to the assumptions of most existing Electromigration (EM) analysis tools, the evolution of EM-induced stress is inherently non-deterministic, influenced by factors such as input current fluctuations and manufacturing non-idealities. Traditional approaches for estimating stress variations typically involve computationally expensive and inefficient Monte Carlo simulations with industrial solvers, which quantify variations using mean and variance metrics. In this work, we introduce a novel machine learning-based framework, termed BPINN-EM- Post, for efficient stochastic analysis of EM-induced post-voiding aging processes. For the first time, our new approach integrates closed-form analytical solutions with a Bayesian Physics- Informed Neural Network (BPINN) framework to accelerate the analysis. The closed-form solutions enforce physical laws at the individual wire segment level, while the BPINN ensures that physics constraints at inter-segment junctions are satisfied and stochastic behaviors are accurately modeled. By reducing the number of variables in the loss functions through utilizing analytical solutions, our method significantly improves training efficiency without accuracy loss and naturally incorporates variational effects. Additionally, the analytical solutions effectively address the challenge of incorporating initial stress distributions in interconnect structures during post-void stress calculations. Numerical results demonstrate that BPINN-EM-Post achieves over 240x and more than 67x speedup compared to Monte Carlo simulations using the FEM-based COMSOL solver and FDM-based EMSpice, respectively, with marginal accuracy loss.

cs.LG

Enhanced Hybrid Temporal Computing Using Deterministic Summations for Ultra-Low-Power Accelerators

This paper presents an accuracy-enhanced Hybrid Temporal Computing (E-HTC) framework for ultra-low-power hardware accelerators with deterministic additions. Inspired by the recently proposed HTC architecture, which leverages pulse-rate and temporal data encoding to reduce switching activity and energy consumption but loses accuracy due to its multiplexer (MUX)-based scaled addition, we propose two bitstream addition schemes: (1) an Exact Multiple-input Binary Accumulator (EMBA), which performs precise binary accumulation, and (2) a Deterministic Threshold-based Scaled Adder (DTSA), which employs threshold logic for scaled addition. These adders are integrated into a multiplier accumulator (MAC) unit supporting both unipolar and bipolar encodings. To validate the framework, we implement two accelerators: a Finite Impulse Response (FIR) filter and an 8-point Discrete Cosine Transform (DCT)/iDCT engine. Results on a 4x4 MAC show that, in unipolar mode, E-HTC matches the RMSE of state-of-the-art Counter-Based Stochastic Computing (CBSC) MAC, improves accuracy by 94% over MUX-based HTC, and reduces power and area by 23% and 7% compared to MUX-based HTC and 64% and 74% compared to CBSC. In bipolar mode, E-HTC MAC achieves 2.09% RMSE -- an 83% improvement over MUX-based HTC -- and approaches CBSC's 1.40% RMSE with area and power savings of 28% and 43% vs. MUX-based HTC and about 76% vs. CBSC. In FIR experiments, both E-HTC variants yield PSNR gains of 3--5 dB (30--45% RMSE reduction) while saving 13% power and 3% area. For DCT/iDCT, E-HTC boosts PSNR by 10--13 dB (70--75% RMSE reduction) while saving area and power over both MUX- and CBSC-based designs.

cs.AR

EMSpice 2.1: A Coupled EM and IR Drop Analysis Tool with Joule Heating and Thermal Map Integration for VLSI Reliability

Electromigration (EM) remains a critical reliability concern in current and future copper-based VLSI circuits. As technology scales down, EM-induced IR drop becomes increasingly severe. While several EM-aware IR drop analysis tools have been proposed, few incorporate the real impact of temperature distribution on both EM and IR drop effects. In this work, we introduce EMSpice 2.1, an enhanced tool built upon the existing coupled IR-EM analysis framework, EMSpice 2.0, for EM-aware IR drop analysis. For the first time, EMSpice 2.1 uniquely integrates Joule heating effects and practical thermal maps derived from actual chip conditions. Additionally, it features improved interoperability with commercial EDA tools, facilitating more comprehensive EM and IR drop sign-off analysis. Our findings demonstrate that specific hotspot patterns significantly impact the lifetime of interconnects and overall chip reliability due to EM failures. Furthermore, our tool exhibits strong agreement with industry-standard tools such as COMSOL, achieving a speedup of over 200 times while maintaining high accuracy.

eess.SY

Hybrid Temporal Computing for Lower Power Hardware Accelerators

In this paper, we propose a new hybrid temporal computing (HTC) framework that leverages both pulse rate and temporal data encoding to design ultra-low energy hardware accelerators. Our approach is inspired by the recently proposed temporal computing, or race logic, which encodes data values as single delays, leading to significantly lower energy consumption due to minimized signal switching. However, race logic is limited in its applications due to inherent restrictions. The new HTC framework overcomes these limitations by encoding signals in both temporal and pulse rate formats for multiplication and in temporal format for propagation. This approach maintains reduced switch energy while being general enough to implement a wide range of arithmetic operations. We demonstrate how HTC multiplication is performed for both unipolar and bipolar data encoding and present the basic designs for multipliers, adders, and MAC units. Additionally, we implement two hardware accelerators: a Finite Impulse Response (FIR) filter and a Discrete Cosine Transform (DCT)/iDCT engine for image compression and DSP applications. Experimental results show that the HTC MAC has a significantly smaller power and area footprint compared to the Unary MAC design and is orders of magnitude faster. Compared to the CBSC MAC, the HTC MAC reduces power consumption by $45.2\%$ and area footprint by $50.13\%$. For the FIR design, the HTC design significantly outperforms the Unary design on all metrics. Compared to the CBSC design, the HTC-based FIR filter reduces power consumption by $36.61\%$ and area cost by $45.85\%$. The HTC-based DCT filter retains the quality of the original image with a decent PSNR, while consuming $23.34\%$ less power and occupying $18.20\%$ less area than the CBSC MAC-based DCT filter.

cs.AR

Run-Time Accuracy Reconfigurable Stochastic Computing for Dynamic Reliability and Power Management

In this paper, we propose a novel accuracy-reconfigurable stochastic computing (ARSC) framework for dynamic reliability and power management. Different than the existing stochastic computing works, where the accuracy versus power/energy trade-off is carried out in the design time, the new ARSC design can change accuracy or bit-width of the data in the run-time so that it can accommodate the long-term aging effects by slowing the system clock frequency at the cost of accuracy while maintaining the throughput of the computing. We validate the ARSC concept on a discrete cosine transformation (DCT) and inverse DCT designs for image compressing/decompressing applications, which are implemented on Xilinx Spartan-6 family XC6SLX45 platform. Experimental results shows that the new design can easily mitigate the long-term aging induced effects by accuracy trade-off while maintaining the throughput of the whole computing process using simple frequency scaling. We further show that one-bit precision loss for input data, which translated to 3.44dB of the accuracy loss in term of Peak Signal to Noise Ratio for images, we can sufficiently compensate the NBTI induced aging effects in 10 years while maintaining the pre-aging computing throughput of 7.19 frames per second. At the same time, we can save 74\% power consumption by 10.67dB of accuracy loss. The proposed ARSC computing framework also allows much aggressive frequency scaling, which can lead to order of magnitude power savings compared to the traditional dynamic voltage and frequency scaling (DVFS) techniques.

cs.AR

EM-GAN: Fast Stress Analysis for Multi-Segment Interconnect Using Generative Adversarial Networks

In this paper, we propose a fast transient hydrostatic stress analysis for electromigration (EM) failure assessment for multi-segment interconnects using generative adversarial networks (GANs). Our work leverages the image synthesis feature of GAN-based generative deep neural networks. The stress evaluation of multi-segment interconnects, modeled by partial differential equations, can be viewed as time-varying 2D-images-to-image problem where the input is the multi-segment interconnects topology with current densities and the output is the EM stress distribution in those wire segments at the given aging time. Based on this observation, we train conditional GAN model using the images of many self-generated multi-segment wires and wire current densities and aging time (as conditions) against the COMSOL simulation results. Different hyperparameters of GAN were studied and compared. The proposed algorithm, called {\it EM-GAN}, can quickly give accurate stress distribution of a general multi-segment wire tree for a given aging time, which is important for full-chip fast EM failure assessment. Our experimental results show that the EM-GAN shows 6.6\% averaged error compared to COMSOL simulation results with orders of magnitude speedup. It also delivers 8.3X speedup over state-of-the-art analytic based EM analysis solver.

cs.LG

GLU3.0: Fast GPU-based Parallel Sparse LU Factorization for Circuit Simulation

LU factorization for sparse matrices is the most important computing step for many engineering and scientific computing problems such as circuit simulation. But parallelizing LU factorization with the Graphic Processing Units (GPU) still remains a challenging problem due to high data dependency and irregular memory accesses. Recently GPU-based hybrid right-looking sparse LU solver, called GLU (1.0 and 2.0), has been proposed to exploit the fine grain level parallelism of GPU. However, a new type of data dependency (called double-U dependency) introduced by GLU slows down the preprocessing step. Furthermore, GLU uses fixed GPU thread allocation strategy, which limits the parallelism. In this article, we propose a new GPU-based sparse LU factorization method, called {\it GLU3.0}, which solves the aforementioned problems. First, it introduces a much more efficient data dependency detection algorithm. Second, we observe that the potential parallelism is different as the matrix factorization goes on. We then develop three different modes of GPU kernel which adapt to different stages to accommodate the computing task changes in the factorization. Experimental results on circuit matrices from University of Florida Sparse Matrix Collection (UFL) show that GLU3.0 delivers 2-3 orders of magnitude speedup over GLU2.0 for the data dependency detection. Furthermore, GLU3.0 achieve 13.0 $\times$ (arithmetic mean) or 6.7$\times$ (geometric mean) speedup over GLU2.0 and 7.1$\times$ (arithmetic mean) or 4.8 $\times$ (geometric mean) over the recently proposed enhanced GLU2.0 sparse LU solver on the same set of circuit matrices.

cs.DC

GPU-based Ising Computing for Solving Balanced Min-Cut Graph Partitioning Problem

Ising computing provides a new computing paradigm for many hard combinatorial optimization problems. Ising computing essentially tries to solve the quadratic unconstrained binary optimization problem, which is also described by the Ising spin glass model and is also the basis for so-called Quantum Annealing computers. In this work, we propose a novel General Purpose Graphics Processing Unit (GPGPU) solver for the balanced min-cut graph partitioning problem, which has many applications in the area of design automation and others. Ising model solvers for the balanced min-cut partitioning problem have been proposed in the past. However, they have rarely been demonstrated in existing quantum computers for many meaningful problem sizes. One difficulty is the fact that the balancing constraint in the balanced min-cut problem can result in a complete graph in the Ising model, which makes each local update a global update. Such global update from each GPU thread will diminish the efficiency of GPU computing, which favors many localized memory accesses for each thread. To mitigate this problem, we propose an novel Global Decoupled Ising (GDI) model and the corresponding annealing algorithm, in which the local update is still preserved to maintain the efficiency. As a result, the new Ising solver essentially eliminates the need for the fully connected graph and will use a more efficient method to track and update global balance without sacrificing cut quality. Experimental results show that the proposed Ising-based min-cut partitioning method outperforms the state of art partitioning tool, METIS, on G-set graph benchmarks in terms of partitioning quality with similar CPU/GPU times.

cs.ET

GPU Based Parallel Ising Computing for Combinatorial Optimization Problems in VLSI Physical Design

In VLSI physical design, many algorithms require the solution of difficult combinatorial optimization problems such as max/min-cut, max-flow problems etc. Due to the vast number of elements typically found in this problem domain, these problems are computationally intractable leading to the use of approximate solutions. In this work, we explore the Ising spin glass model as a solution methodology for hard combinatorial optimization problems using the general purpose GPU (GPGPU). The Ising model is a mathematical model of ferromagnetism in statistical mechanics. Ising computing finds a minimum energy state for the Ising model which essentially corresponds to the expected optimal solution of the original problem. Many combinatorial optimization problems can be mapped into the Ising model. In our work, we focus on the max-cut problem as it is relevant to many VLSI physical design problems. Our method is inspired by the observation that Ising annealing process is very amenable to fine-grain massive parallel GPU computing. We will illustrate how the natural randomness of GPU thread scheduling can be exploited during the annealing process to create random update patterns and allow better GPU resource utilization. Furthermore, the proposed GPU-based Ising computing can handle any general Ising graph with arbitrary connections, which was shown to be difficult for existing FPGA and other hardware based implementation methods. Numerical results show that the proposed GPU Ising max-cut solver can deliver more than 2000X speedup over the CPU version of the algorithm on some large examples, which shows huge performance improvement for addressing many hard optimization algorithms for practical VLSI physical design.

physics.comp-ph

DEEPEYE: A Compact and Accurate Video Comprehension at Terminal Devices Compressed with Quantization and Tensorization

As it requires a huge number of parameters when exposed to high dimensional inputs in video detection and classification, there is a grand challenge to develop a compact yet accurate video comprehension at terminal devices. Current works focus on optimizations of video detection and classification in a separated fashion. In this paper, we introduce a video comprehension (object detection and action recognition) system for terminal devices, namely DEEPEYE. Based on You Only Look Once (YOLO), we have developed an 8-bit quantization method when training YOLO; and also developed a tensorized-compression method of Recurrent Neural Network (RNN) composed of features extracted from YOLO. The developed quantization and tensorization can significantly compress the original network model yet with maintained accuracy. Using the challenging video datasets: MOMENTS and UCF11 as benchmarks, the results show that the proposed DEEPEYE achieves 3.994x model compression rate with only 0.47% mAP decreased; and 15,047x parameter reduction and 2.87x speed-up with 16.58% accuracy improvement.

cs.CV