arXiv subjects

Rebecca Pelke

Publications and source records attributed to Rebecca Pelke.

15 recordsLinked to original sources

arXiv2026

Optimizing Binary and Ternary Neural Network Inference on RRAM Crossbars using CIM-Explorer

Using Resistive Random Access Memory (RRAM) crossbars in Computing-in-Memory (CIM) architectures offers a promising solution to overcome the von Neumann bottleneck. Due to non-idealities like cell variability, RRAM crossbars are often operated in binary mode, utilizing only two states: Low Resistive State (LRS) and High Resistive State (HRS). Binary Neural Networks (BNNs) and Ternary Neural Networks (TNNs) are well-suited for this hardware due to their efficient mapping. Existing software projects for RRAM-based CIM typically focus on only one aspect: compilation, simulation, or Design Space Exploration (DSE). Moreover, they often rely on classical 8 bit quantization. To address these limitations, we introduce CIM-Explorer, a modular toolkit for optimizing BNN and TNN inference on RRAM crossbars. CIM-Explorer includes an end-to-end compiler stack, multiple mapping options, and simulators, enabling a DSE flow for accuracy estimation across different crossbar parameters and mappings. CIM-Explorer can accompany the entire design process, from early accuracy estimation for specific crossbar parameters, to selecting an appropriate mapping, and compiling BNNs and TNNs for a finalized crossbar chip. In DSE case studies, we demonstrate the expected accuracy for various mappings and crossbar parameters. CIM-Explorer can be found on GitHub.

cs.ET↗

arXiv2026

Optimizing ML Workload Partitioning between CPUs and CIM Accelerators for Heterogeneous Computing

Computing-in-Memory (CIM) accelerators execute Matrix-Vector Multiplications (MVMs) in memory, making them a compelling solution for Machine Learning (ML) workloads. However, existing ML workload partitioning approaches for CIM accelerators do not fully account for Resistive Random Access Memory (RRAM) constraints such as limited memory, high write latency, and limited endurance. They also neglect parallelism, low-level architectural effects, or the Central Processing Unit (CPU) as a complementary compute resource. To address these limitations, we propose an Integer Linear Programming (ILP)-based workload partitioning framework for heterogeneous CPU-CIM systems. It minimizes end-to-end inference latency under RRAM constraints, captures parallelism, and combines empirical profiling with analytical models. Using our framework, heterogeneous CPU-CIM execution achieves speedups of up to 30.9x over CPU-only execution on an edge CPU and 7.3x over a high-performance CPU. A Design Space Exploration (DSE) yields further design insights for future CIM accelerators.

cs.ET↗

arXiv2026

Mixed-Precision Training and Compilation for RRAM-based Computing-in-Memory Accelerators

Computing-in-Memory (CIM) accelerators are a promising solution for accelerating Machine Learning (ML) workloads, as they perform Matrix-Vector Multiplications (MVMs) on crossbar arrays directly in memory. Although the bit widths of the crossbar inputs and cells are very limited, most CIM compilers do not support quantization below 8 bit. As a result, a single MVM requires many compute cycles, and weights cannot be efficiently stored in a single crossbar cell. To address this problem, we propose a mixed-precision training and compilation framework for CIM architectures. The biggest challenge is the massive search space, that makes it difficult to find good quantization parameters. This is why we introduce a reinforcement learning-based strategy to find suitable quantization configurations that balance latency and accuracy. In the best case, our approach achieves up to a 2.48x speedup over existing state-of-the-art solutions, with an accuracy loss of only 0.086 %.

cs.LG↗

arXiv2026

Introducing Instruction-Accurate Simulators for Performance Estimation of Autotuning Workloads

Accelerating Machine Learning (ML) workloads requires efficient methods due to their large optimization space. Autotuning has emerged as an effective approach for systematically evaluating variations of implementations. Traditionally, autotuning requires the workloads to be executed on the target hardware (HW). We present an interface that allows executing autotuning workloads on simulators. This approach offers high scalability when the availability of the target HW is limited, as many simulations can be run in parallel on any accessible HW. Additionally, we evaluate the feasibility of using fast instruction-accurate simulators for autotuning. We train various predictors to forecast the performance of ML workload implementations on the target HW based on simulation statistics. Our results demonstrate that the tuned predictors are highly effective. The best workload implementation in terms of actual run time on the target HW is always within the top 3 % of predictions for the tested x86, ARM, and RISC-V-based architectures. In the best case, this approach outperforms native execution on the target HW for embedded architectures when running as few as three samples on three simulators in parallel.

cs.AR↗

arXiv2026

NQC2: A Non-Intrusive QEMU Code Coverage Plugin

Code coverage analysis has become a standard approach in software development, facilitating the assessment of test suite effectiveness, the identification of under-tested code segments, and the discovery of performance bottlenecks. When code coverage of software for embedded systems needs to be measured, conventional approaches quickly meet their limits. A commonly used approach involves instrumenting the source files with added code that collects and dumps coverage information during runtime. This inserted code usually relies on the existence of an operating and a file system to dump the collected data. These features are not available for bare-metal programs that are executed on embedded systems. To overcome this issue, we present NQC2, a plugin for QEMU.NQC2 extracts coverage information from QEMU during runtime and stores them into a file on the host machine. This approach is even compatible with modified QEMU versions and does not require target-software instrumentation. NQC2 outperforms a comparable approach from Xilinx by up to 8.5 x.

cs.SE↗

arXiv2025

Evaluating the Scalability of Binary and Ternary CNN Workloads on RRAM-based Compute-in-Memory Accelerators

The increasing computational demand of Convolutional Neural Networks (CNNs) necessitates energy-efficient acceleration strategies. Compute-in-Memory (CIM) architectures based on Resistive Random Access Memory (RRAM) offer a promising solution by reducing data movement and enabling low-power in-situ computations. However, their efficiency is limited by the high cost of peripheral circuits, particularly Analog-to-Digital Converters (ADCs). Large crossbars and low ADC resolutions are often used to mitigate this, potentially compromising accuracy. This work introduces novel simulation methods to model the impact of resistive wire parasitics and limited ADC resolution on RRAM crossbars. Our parasitics model employs a vectorised algorithm to compute crossbar output currents with errors below 0.15% compared to SPICE. Additionally, we propose a variable step-size ADC and a calibration methodology that significantly reduces ADC resolution requirements. These accuracy models are integrated with a statistics-based energy model. Using our framework, we conduct a comparative analysis of binary and ternary CNNs. Experimental results demonstrate that the ternary CNNs exhibit greater resilience to wire parasitics and lower ADC resolution but suffer a 40% reduction in energy efficiency. These findings provide valuable insights for optimising RRAM-based CIM accelerators for energy-efficient deep learning.

eess.SP↗

arXiv2025

High-Performance ARM-on-ARM Virtualization for Multicore SystemC-TLM-Based Virtual Platforms

The increasing complexity of hardware and software requires advanced development and test methodologies for modern systems on chips. This paper presents a novel approach to ARM-on-ARM virtualization within SystemC-based simulators using Linux's KVM to achieve high-performance simulation. By running target software natively on ARM-based hosts with hardware-based virtualization extensions, our method eliminates the need for instruction-set simulators, which significantly improves performance. We present a multicore SystemC-TLM-based CPU model that can be used as a drop-in replacement for an instruction-set simulator. It places no special requirements on the host system, making it compatible with various environments. Benchmark results show that our ARM-on-ARM-based virtual platform achieves up to 10 x speedup over traditional instruction-set-simulator-based models on compute-intensive workloads. Depending on the benchmark, speedups increase to more than 100 x.

cs.SE↗

arXiv2025

Bridging the Gap: Physical PCI Device Integration Into SystemC-TLM Virtual Platforms

In today's technology-driven world, early-stage software development and testing are crucial. Virtual Platforms (VPs) have become indispensable tools for this purpose as they serve as a platform to execute and debug the unmodified target software at an early design stage. With the increasing complexity of software, especially in areas like Artificial Intelligence (AI) applications, VPs need to provide high simulation speed to ensure the target software executes within a reasonable time. Hybrid simulation, which combines virtual models with real hardware, can improve the performance of VPs. This paper introduces a novel approach for integrating real Peripheral Component Interconnect (PCI) devices into SystemC-TLM-2.0-based VPs. The embedded PCI devices enable high performance, easy integration, and allow introspection for analysis and optimization. To illustrate the practical application of our approach, we present a case study where we integrate Google Coral's Edge Tensor Processing Unit (TPU) into an ARM-based VP. The integration allows efficient execution of AI workloads, accelerating simulation speeds by up to 480x while eliminating the need for complex virtual device models. Beyond accelerating AI-workload execution, our framework enables driver development, regression testing across architectures, and device communication analysis. Our findings demonstrate that embedding PCI devices into SystemC simulations significantly enhances

cs.SE↗

arXiv2024

A Calibratable Model for Fast Energy Estimation of MVM Operations on RRAM Crossbars

The surge in AI usage demands innovative power reduction strategies. Novel Compute-in-Memory (CIM) architectures, leveraging advanced memory technologies, hold the potential for significantly lowering energy consumption by integrating storage with parallel Matrix-Vector-Multiplications (MVMs). This study addresses the 1T1R RRAM crossbar, a core component in numerous CIM architectures. We introduce an abstract model and a calibration methodology for estimating operational energy. Our tool condenses circuit-level behaviour into a few parameters, facilitating energy assessments for DNN workloads. Validation against low-level SPICE simulations demonstrates speedups of up to 1000x and energy estimations with errors below 1%.

eess.SP↗

arXiv2024

A Fully Automated Platform for Evaluating ReRAM Crossbars

Resistive Random Access Memory (ReRAM) is a promising candidate for implementing Computing-in-Memory (CIM) architectures and neuromorphic circuits. ReRAM cells exhibit significant variability across different memristive devices and cycles, necessitating further improvements in the areas of devices, algorithms, and applications. To achieve this, understanding the stochastic behavior of the different ReRAM technologies is essential. The NeuroBreakoutBoard (NBB) is a versatile instrumentation platform to characterize Non-Volatile Memories (NVMs). However, the NBB itself does not provide any functionality in the form of software or a controller. In this paper, we present a control board for the NBB able to perform reliability assessments of 1T1R ReRAM crossbars. In more detail, an interface that allows a host PC to communicate with the NBB via the new control board is implemented. In a case study, we analyze the Cycle-to-Cycle (C2C) variation and read disturb TiN/Ti/HfO2/TiN cells for different read voltages to gain an understanding of their operational behavior.

cs.ET↗

arXiv2024

QTFlow: Quantitative Timing-Sensitive Information Flow for Security-Aware Hardware Design on RTL

In contemporary Electronic Design Automation (EDA) tools, security often takes a backseat to the primary goals of power, performance, and area optimization. Commonly, the security analysis is conducted by hand, leading to vulnerabilities in the design remaining unnoticed. Security-aware EDA tools assist the designer in the identification and removal of security threats while keeping performance and area in mind. Cutting-edge methods employ information flow analysis to identify inadvertent information leaks in design structures. Current information leakage detection methods use quantitative information flow analysis to quantify the leaks. However, handling sequential circuits poses challenges for state-of-the-art techniques due to their time-agnostic nature, overlooking timing channels, and introducing false positives. To address this, we introduce QTFlow, a timing-sensitive framework for quantifying hardware information leakages during the design phase. Illustrating its effectiveness on open-source benchmarks, QTFlow autonomously identifies timing channels and diminishes all false positives arising from time-agnostic analysis when contrasted with current state-of-the-art techniques.

cs.CR↗

arXiv2024

CLSA-CIM: A Cross-Layer Scheduling Approach for Computing-in-Memory Architectures

The demand for efficient machine learning (ML) accelerators is growing rapidly, driving the development of novel computing concepts such as resistive random access memory (RRAM)-based tiled computing-in-memory (CIM) architectures. CIM allows to compute within the memory unit, resulting in faster data processing and reduced power consumption. Efficient compiler algorithms are essential to exploit the potential of tiled CIM architectures. While conventional ML compilers focus on code generation for CPUs, GPUs, and other von Neumann architectures, adaptations are needed to cover CIM architectures. Cross-layer scheduling is a promising approach, as it enhances the utilization of CIM cores, thereby accelerating computations. Although similar concepts are implicitly used in previous work, there is a lack of clear and quantifiable algorithmic definitions for cross-layer scheduling for tiled CIM architectures. To close this gap, we present CLSA-CIM, a cross-layer scheduling algorithm for tiled CIM architectures. We integrate CLSA-CIM with existing weight-mapping strategies and compare performance against state-of-the-art (SOTA) scheduling algorithms. CLSA-CIM improves the utilization by up to 17.9 x , resulting in an overall speedup increase of up to 29.2 x compared to SOTA.

cs.AR↗

arXiv2023

Mapping of CNNs on multi-core RRAM-based CIM architectures

RRAM-based multi-core systems improve the energy efficiency and performance of CNNs. Thereby, the distributed parallel execution of convolutional layers causes critical data dependencies that limit the potential speedup. This paper presents synchronization techniques for parallel inference of convolutional layers on RRAM-based CIM architectures. We propose an architecture optimization that enables efficient data exchange and discuss the impact of different architecture setups on the performance. The corresponding compiler algorithms are optimized for high speedup and low memory consumption during CNN inference. We achieve more than 99% of the theoretical acceleration limit with a marginal data transmission overhead of less than 4% for state-of-the-art CNN benchmarks.

cs.AR↗

arXiv2023

Work-in-Progress: A Universal Instrumentation Platform for Non-Volatile Memories

Emerging non-volatile memories (NVMs) represent a disruptive technology that allows a paradigm shift from the conventional von Neumann architecture towards more efficient computing-in-memory (CIM) architectures. Several instrumentation platforms have been proposed to interface NVMs allowing the characterization of single cells and crossbar structures. However, these platforms suffer from low flexibility and are not capable of performing CIM operations on NVMs. Therefore, we recently designed and built the NeuroBreakoutBoard, a highly versatile instrumentation platform capable of executing CIM on NVMs. We present our preliminary results demonstrating a relative error < 5% in the range of 1 k$Ω$ to 1 M$Ω$ and showcase the switching behavior of a HfO$_2$/Ti-based memristive cell.

cs.AR↗

arXiv2023

Fault Injection in Native Logic-in-Memory Computation on Neuromorphic Hardware

Logic-in-memory (LIM) describes the execution of logic gates within memristive crossbar structures, promising to improve performance and energy efficiency. Utilizing only binary values, LIM particularly excels in accelerating binary neural networks, shifting it in the focus of edge applications. Considering its potential, the impact of faults on BNNs accelerated with LIM still lacks investigation. In this paper, we propose faulty logic-in-memory (FLIM), a fault injection platform capable of executing full-fledged BNNs on LIM while injecting in-field faults. The results show that FLIM runs a single MNIST picture 66754x faster than the state of the art by offering a fine-grained fault injection methodology.

cs.ET↗