SEARCH · Engineering Papers
Results for “non-volatile memory (NVM)”
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Design-Technology Co-Optimization for NVM-based Neuromorphic Processing Elements
An emerging use-case of machine learning (ML) is to train a model on a high-performance system and deploy the trained model on energy-constrained embedded systems. Neuromorphic hardware platforms, which operate on principles of the biological brain, can significantly lower the energy overhead of a machine learning inference task, making these platforms an attractive solution for embedded ML systems. In this paper, we present a design-technology tradeoff analysis to implement such inference tasks on the processing elements (PEs) of a Non-Volatile Memory (NVM)-based neuromorphic hardware. Through detailed circuit-level simulations at scaled process technology nodes, we show the negative impact of technology scaling on the information-processing latency, which impacts the quality-of-service (QoS) of an embedded ML system. At a finer granularity, the latency inside a PE depends on 1) the delay introduced by parasitic components on its current paths, and 2) the varying delay to sense different resistance states of its NVM cells. Based on these two observations, we make the following three contributions. First, on the technology front, we propose an optimization scheme where the NVM resistance state that takes the longest time to sense is set on current paths having the least delay, and vice versa, reducing the average PE latency, which improves the QoS. Second, on the architecture front, we introduce isolation transistors within each PE to partition it into regions that can be individually power-gated, reducing both latency and energy. Finally, on the system-software front, we propose a mechanism to leverage the proposed technological and architectural enhancements when implementing a machine-learning inference task on neuromorphic PEs of the hardware. Evaluations with a recent neuromorphic hardware architecture show that our proposed design-technology co-optimization approach improves both performance and energy efficiency of machine-learning inference tasks without incurring high cost-per-bit.
Requirements and Usage of NVM in Advanced Onboard Data Processing Systems
This viewgraph presentation gives an overview of the requirements and uses of non-volatile memory (NVM) in advanced onboard data processing systems. Supercomputing in space presents the only viable approach to the bandwidth problem (can't get data down to Earth), controlling constellations of cooperating satellites, reducing mission operating costs, and real-time intelligent decision making and science data gathering. Details are given on the REE vision and impact on NASA and Department of Defense missions, objectives of REE, baseline architecture, and issues. NVM uses and requirements are listed.
Progress and Challenges for Memtransistors in Neuromorphic Circuits and Systems
Due to the increasing importance of artificial intelligence (AI), significant recent effort has been devoted to the development of neuromorphic circuits that seek to emulate the energy-efficient information processing of the brain. While non-volatile memory (NVM) based on resistive switches, phase-change memory, and magnetic tunnel junctions has shown potential for implementing neural networks, additional multi-terminal device concepts are required for more sophisticated bio-realistic functions. Of particular interest are memtransistors based on low-dimensional nanomaterials, which are capable of electrostatically tuning memory and learning behavior at the device level. Herein, a conceptual overview of the memtransistor is provided in the context of neuromorphic circuits. Recent progress is surveyed for memtransistors and related multi-terminal NVM devices including dual-gated floating-gate memories, dual-gated ferroelectric transistors, and dual-gated van der Waals heterojunctions. Further, the different materials systems and device architectures are classified based on the degree of control and relative tunability of synaptic behavior, with an emphasis on device concepts that harness the reduced dimensionality, weak electrostatic screening, and phase-changes properties of nanomaterials. Finally, strategies for achieving wafer-scale integration of memtransistors and multi-terminal NVM devices are delineated, with specific attention given to the materials challenges for practical neuromorphic circuits.
Introduction: Neuromorphic Materials
The explosive growth in data collection and the need to process it efficiently, as well as the desire to automate increasingly complex tasks in transportation, medical care, manufacturing, security and many other fields have motivated a growing interest in neuromorphic computing. Unlike the binary, transistorbased ON/OFF logic gates and separate logic and memory functionalities employed in digital computing, neuromorphic computing is inspired by animal brains that use interconnected synapses and neurons to perform processing, storage and transmission of information at the same location, while only consuming ~20 W or less of power. Motivated by the brain’s efficiency, adaptability, self-learning and resiliency qualities, neuromorphic computing can be broadly defined as an approach to processing and storing information using hardware and algorithms inspired by models of biological neural systems. Present research in neuromorphic computing encompasses approaches that vary significantly in their degree of neuro-inspiration, from systems that only incorporate features such as asynchronous, event-driven operation or use crossbar arrays of non-volatile memory (NVM) elements to accelerate deep neural networks (DNNs), to designs that embrace the extreme parallelism, sparsity, reconfigurability, adaptability, complexity and stochasticity observed in nervous systems. The term ‘neuromorphic’ computing is often credited to Carver Mead, who in the 1980s investigated Si-based analog electronics to replicate functions of the animal retina. Earlier important advances in this field include the work of Frank Rosenblatt, who proposed the concept of the perceptron, Bernard Widrow, who used this concept to build one of the first analog neural networks, the Adaline and many other researchers (see ref. 6 for an historical perspective on neuromorphic computing). With the recent increase in the use of artificial intelligence and large language models, and rising concerns over the associated energy costs, interest in neuromorphic hardware has expanded rapidly. According to some estimates, driven largely by the drastic growth in the training use of artificial intelligence (AI) models using the current computing architectures, the energy cost of computing is projected to reach the energy supply worldwide by 2045. Furthermore, while this is not a realistic outcome, it means that, if more efficient computing technologies are not developed -- soon -- the world will soon become one where demand for energy and market constraints limit the continued increase of societal access to AI and cloud services from data centers. Data centers used for training and use of these models consume hundreds of terawatt hours of electricity, already past 4% of the US electricity demand.
First-principles investigation of the resistive switching energetics in monolayer MoS 2 : insights into metal diffusion and adsorption
A deeper understanding of resistive switching (RS) in 2D materials is essential for advancing neuromorphic computing. The Dissociation-Diffusion-Adsorption (DDA) model offers a useful framework for probing RS mechanisms in non-volatile memory (NVM) and in-memory computing. We have employed first-principles density functional theory (DFT) to explore dissociation, diffusion, and adsorption phenomena within the DDA model, focusing on the interactions between exemplary metal atoms (Au, Ag, Cu) and monolayer MoS 2 . Nudged elastic band (NEB) calculations evaluated diffusion barriers in pristine and sulfur-vacancy MoS 2 . Charged systems were modeled to assess the impact of applied bias on migration pathways. We also examined metal dissociation from bulk electrodes and adsorption at S vacancies. Ag/MoS 2 shows the lowest dissociation barrier (~0.034 eV), while Au and Cu exhibit similar values (~0.32 eV). These insights highlight Ag as a promising candidate for low-energy RS applications and provide guidance for optimizing switching efficiency in 2D memory devices.
UMap: An application-oriented user level memory mapping library
Exploiting the prominent role of complex memories in exascale node architecture, the UMap page fault handler offers new capabilities to access large memory-mapped data sets directly. UMap provides flexible configuration options to customize page handling to each application, including analysis of massive observational and simulation data sets. The high-performance design features I/O decoupling, dynamic load balancing, and application-level controls. Page faults triggered by application threads and processes accessing data mapped to a UMapp’ed region are handled via the Linux userfaultfd protocol, an asynchronous message-oriented kernel-user communication mechanism that avoids the context switch penalty of traditional signal fault handlers. UMap is fully open source. In this paper, we give an overview of the UMap library architecture, its extensible plugin architecture, and the use/performance of UMap in emerging heterogeneous memory hierarchies such as near-node Non-volatile Memory (NVM) and network attached memories. We highlight new capabilities in two pagefault management plugins, the NetworkStore and SparseStore. We demonstrate the integration between UMap and multiple ECP products including Caliper, Metall, ZFP, Mochi, and Ripples.
Inadvertently programmed bits in Samsung 128 Mbit flash devices: a flaky investigation
JPL's X2000 avionics design pioneers new territory by specifying a non-volatile memory (NVM) board based on flash memories. The Samsung 128Mb device chosen was found to demonstrate bit errors (mostly program disturbs) and block-erase failures that increase with cycling. Low temperature, certain pseudo- random patterns, and, probably, higher bias increase the observable bit errors. An experiment was conducted to determine the wearout dependence of the bit errors to 100k cycles at cold temperature using flight-lot devices (some pre-irradiated). The results show an exponential growth rate, a wide part-to-part variation, and some annealing behavior.
Vorago RH-OBC-1 Single Event Effect Characterization Test Report
The Vorago Technologies RH-OBC-1 is a CubeSat Kit Bus compatible single board computer with a Vorago VA10820 ARM Cortex-M0 microcontroller at its core. The board also includes a set of common peripheral integrated circuits, like voltage regulators, non-volatile memories (NVM), an analog-to-digital converter (ADC), a watchdog/supervisor, and a Controller Area Network (CAN) bus transceiver. High-energy proton (200 MeV) testing was conducted at both board and component levels to investigate single-event effects.
NEPP non-volatile memory : FY09 summary report
This report documents the activities and results of the fiscal year 2009 (FY09) funding for the NASA Electronic Parts and Packaging (NEPP) program for non-volatile memory (NVM) devices. The FY09 task was divided into two main efforts: 1. Development of in-house measurement capability for flash devices 2. Reliability characterization of multi-level vs. single-level flash devices The FY09 NEPP Non Volatile Memory study was organized into these two sections in recognition of the continued and increasing importance of NVM to NASA. All NASA missions use NVM for boot code storage and some limited data archiving. Space-grade NVM has lagged commercial NVM development by many orders of magnitude, however. This gap in density only continues to increase as space-grade memories remain constant in the 1–16 Mbit levels, while commercial flash-based devices shrink in size and double in capacity every 18 months. Commercial flash devices are now at the 32 Gbit level, 10,000 times denser than current NASA NVM. This FY marks the beginning of a significant technological transition in NVM for NASA. The aggressive scaling of commercial NVMs has, as a by-product, made some of the flash devices radiation-tolerant enough to be at least considered for low-level radiation environment missions [Irom 2008]. This NEPP task is focused on understanding the reliability and potential future applications of such highly scaled flash memories.
One Time Programmable Antifuse Memory Based on Bulk Junctionless Transistor
One time programmable (OTP) antifuse base memory is demonstrated based on a bulk junctionless gate-all-around (GAA) nanowire transistor technology. The presented memory consists of a single transistor (1T) footprint without any process modification. The source/drain (S/D) and gate respectively become bit line and word line where the antifuse is formed by oxide breakdown across the gate and the channel. The channel is connected directly to the bit line due to junctionless S/D and inherently isolated from the neighboring cell by the GAA channel. Therefore, an array of 1T antifuse OTP can be a candidate for the sub-5-nanometer technology node.
Radiation-Hardened Solid-State Drive
A method is provided for a radiationhardened (rad-hard) solid-state drive for space mission memory applications by combining rad-hard and commercial off-the-shelf (COTS) non-volatile memories (NVMs) into a hybrid architecture. The architecture is controlled by a rad-hard ASIC (application specific integrated circuit) or a FPGA (field programmable gate array). Specific error handling and data management protocols are developed for use in a rad-hard environment. The rad-hard memories are smaller in overall memory density, but are used to control and manage radiation-induced errors in the main, and much larger density, non-rad-hard COTS memory devices. Small amounts of rad-hard memory are used as error buffers and temporary caches for radiation-induced errors in the large COTS memories. The rad-hard ASIC/FPGA implements a variety of error-handling protocols to manage these radiation-induced errors. The large COTS memory is triplicated for protection, and CRC-based counters are calculated for sub-areas in each COTS NVM array. These counters are stored in the rad-hard non-volatile memory. Through monitoring, rewriting, regeneration, triplication, and long-term storage, radiation-induced errors in the large NV memory are managed. The rad-hard ASIC/FPGA also interfaces with the external computer buses.
Non-volatile memory solution for near term NASA missions
Explore the source record for details and available documents.
Neuromorphic Accelerator for Deep Spiking Neural Networks with NVM Crossbar Arrays
In this paper, we present a scalable digital hardware accelerator based on non-volatile memory arrays capable of realizing deep convolutional spiking neural networks (SNNs). Our design studies are conducted using a compact model for spin-transfer torque random access memory (STT-RAM) devices. Large networks are realized by tiling multiple cores which communicate by transmitting spike packets via an on-chip routing network. Compared to an equivalent SRAM based core design, we show that the STT-RAM based design achieves nearly 15X higher GSOPS (Synaptic Operations per Second) per Watt per mm 2 making it a promising platform for realizing systems with significant area and power limitations.
A Radiation-Tolerant, Low-Power Non-Volatile Memory Based on Silicon Nanocrystal Quantum Dots
Nanocrystal nonvolatile floating-gate memories are a good candidate for space applications - initial results suggest they are fast, more reliable and consume less power than conventional floating gate memories. In the nanocrystal based NVM device, charge is not stored on a continuous polysilicon layer (so-called floating gate), but instead on a layer of discrete nanocrystals. Charge injection and storage in dense arrays of silicon nanocrystals in SiO2 is a critical aspect of the performance of potential nanocrystal flash memory structures. The ultimate goal for this class of devices is few- or single-electron storage in a small number of nanocrystal elements. In addition, the nanocrystal layer fabrication technique should be simple, 8-inch wafer compatible and well controlled in program/erase threshold voltage swing was seen during 100,000 program and erase cycles. Additional near-term goals for this project include extensive testing for radiation hardness and the development of artificial layered tunnel barrier heterostructures which have the potential for large speed enhancements for read/write of nanocrystal memory elements, compared with conventional flash devices. Additional information is contained in the original extended abstract.
Embedding security into ferroelectric FET array via in situ memory operation
Non-volatile memories (NVMs) have the potential to reshape next-generation memory systems because of their promising properties of near-zero leakage power consumption, high density and non-volatility. However, NVMs also face critical security threats that exploit the non-volatile property. Compared to volatile memory, the capability of retaining data even after power down makes NVM more vulnerable. Existing solutions to address the security issues of NVMs are mainly based on Advanced Encryption Standard (AES), which incurs significant performance and power overhead. In this paper, we propose a lightweight memory encryption/decryption scheme by exploiting in-situ memory operations with negligible overhead. To validate the feasibility of the encryption/decryption scheme, device-level and array-level experiments are performed using ferroelectric field effect transistor (FeFET) as an example NVM without loss of generality. Besides, a comprehensive evaluation is performed on a 128 × 128 FeFET AND-type memory array in terms of area, latency, power and throughput. Compared with the AES-based scheme, our scheme shows ~22.6×/~14.1× increase in encryption/decryption throughput with negligible power penalty. Furthermore, we evaluate the performance of our scheme over the AES-based scheme when deploying different neural network workloads. Our scheme yields significant latency reduction by 90% on average for encryption and decryption processes.