Showing papers by "Hewlett-Packard published in 2018"

The future of electronics based on memristive systems

Abstract: A memristor is a resistive device with an inherent memory. The theoretical concept of a memristor was connected to physically measured devices in 2008 and since then there has been rapid progress in the development of such devices, leading to a series of recent demonstrations of memristor-based neuromorphic hardware systems. Here, we evaluate the state of the art in memristor-based electronics and explore where the future of the field lies. We highlight three areas of potential technological impact: on-chip memory and storage, biologically inspired computing and general-purpose in-memory computing. We analyse the challenges, and possible solutions, associated with scaling the systems up for practical applications, and consider the benefits of scaling the devices down in terms of geometry and also in terms of obtaining fundamental control of the atomic-level dynamics. Finally, we discuss the ways we believe biology will continue to provide guiding principles for device innovation and system optimization in the field. This Perspective evaluates the state of the art in memristor-based electronics and explores the future development of such devices in on-chip memory, biologically inspired computing and general-purpose in-memory computing.

Analogue signal and image processing with large memristor crossbars

Abstract: Memristor crossbars offer reconfigurable non-volatile resistance states and could remove the speed and energy efficiency bottleneck in vector-matrix multiplication, a core computing task in signal and image processing. Using such systems to multiply an analogue-voltage-amplitude-vector by an analogue-conductance-matrix at a reasonably large scale has, however, proved challenging due to difficulties in device engineering and array integration. Here we show that reconfigurable memristor crossbars composed of hafnium oxide memristors on top of metal-oxide-semiconductor transistors are capable of analogue vector-matrix multiplication with array sizes of up to 128 × 64 cells. Our output precision (5–8 bits, depending on the array size) is the result of high device yield (99.8%) and the multilevel, stable states of the memristors, while the linear device current–voltage characteristics and low wire resistance between cells leads to high accuracy. With the large memristor crossbars, we demonstrate signal processing, image compression and convolutional filtering, which are expected to be important applications in the development of the Internet of Things (IoT) and edge computing.

Fully memristive neural networks for pattern classification with unsupervised learning

Abstract: Neuromorphic computers comprised of artificial neurons and synapses could provide a more efficient approach to implementing neural network algorithms than traditional hardware. Recently, artificial neurons based on memristors have been developed, but with limited bio-realistic dynamics and no direct interaction with the artificial synapses in an integrated network. Here we show that a diffusive memristor based on silver nanoparticles in a dielectric film can be used to create an artificial neuron with stochastic leaky integrate-and-fire dynamics and tunable integration time, which is determined by silver migration alone or its interaction with circuit capacitance. We integrate these neurons with nonvolatile memristive synapses to build fully memristive artificial neural networks. With these integrated networks, we experimentally demonstrate unsupervised synaptic weight updating and pattern classification.

Efficient and self-adaptive in-situ learning in multilayer memristor neural networks

Abstract: Memristors with tunable resistance states are emerging building blocks of artificial neural networks. However, in situ learning on a large-scale multiple-layer memristor network has yet to be demonstrated because of challenges in device property engineering and circuit integration. Here we monolithically integrate hafnium oxide-based memristors with a foundry-made transistor array into a multiple-layer neural network. We experimentally demonstrate in situ learning capability and achieve competitive classification accuracy on a standard machine learning dataset, which further confirms that the training algorithm allows the network to adapt to hardware imperfections. Our simulation using the experimental parameters suggests that a larger network would further increase the classification accuracy. The memristor neural network is a promising hardware platform for artificial intelligence with high speed-energy efficiency.

Memristor-Based Analog Computation and Neural Network Classification with a Dot Product Engine.

Abstract: Using memristor crossbar arrays to accelerate computations is a promising approach to efficiently implement algorithms in deep neural networks. Early demonstrations, however, are limited to simulations or small-scale problems primarily due to materials and device challenges that limit the size of the memristor crossbar arrays that can be reliably programmed to stable and analog values, which is the focus of the current work. High-precision analog tuning and control of memristor cells across a 128 × 64 array is demonstrated, and the resulting vector matrix multiplication (VMM) computing precision is evaluated. Single-layer neural network inference is performed in these arrays, and the performance compared to a digital approach is assessed. Memristor computing system used here reaches a VMM accuracy equivalent of 6 bits, and an 89.9% recognition accuracy is achieved for the 10k MNIST handwritten digit test set. Forecasts show that with integrated (on chip) and scaled memristors, a computational efficiency greater than 100 trillion operations per second per Watt is possible.

An artificial nociceptor based on a diffusive memristor.

Abstract: A nociceptor is a critical and special receptor of a sensory neuron that is able to detect noxious stimulus and provide a rapid warning to the central nervous system to start the motor response in the human body and humanoid robotics. It differs from other common sensory receptors with its key features and functions, including the “no adaptation” and “sensitization” phenomena. In this study, we propose and experimentally demonstrate an artificial nociceptor based on a diffusive memristor with critical dynamics for the first time. Using this artificial nociceptor, we further built an artificial sensory alarm system to experimentally demonstrate the feasibility and simplicity of integrating such novel artificial nociceptor devices in artificial intelligence systems, such as humanoid robots. The development of humanoid robots with artificial intelligence calls for smart solutions for tactile sensing systems that respond to dynamic changes in the environment. Here, Yoon et al. emulate non-adaption and sensitization function of a nociceptor—a sensory neuron—using diffusive oxide-based memristors.

A Manifesto for Future Generation Cloud Computing: Research Directions for the Next Decade

Abstract: The Cloud computing paradigm has revolutionised the computer science horizon during the past decade and has enabled the emergence of computing as the fifth utility. It has captured significant attention of academia, industries, and government bodies. Now, it has emerged as the backbone of modern economy by offering subscription-based services anytime, anywhere following a pay-as-you-go model. This has instigated (1) shorter establishment times for start-ups, (2) creation of scalable global enterprise applications, (3) better cost-to-value associativity for scientific and high-performance computing applications, and (4) different invocation/execution models for pervasive and ubiquitous applications. The recent technological developments and paradigms such as serverless computing, software-defined networking, Internet of Things, and processing at network edge are creating new opportunities for Cloud computing. However, they are also posing several new challenges and creating the need for new approaches and research strategies, as well as the re-evaluation of the models that were developed to address issues such as scalability, elasticity, reliability, security, sustainability, and application models. The proposed manifesto addresses them by identifying the major open challenges in Cloud computing, emerging trends, and impact areas. It then offers research directions for the next decade, thus helping in the realisation of Future Generation Cloud Computing.

Capacitive neural network with neuro-transistors.

Abstract: Experimental demonstration of resistive neural networks has been the recent focus of hardware implementation of neuromorphic computing. Capacitive neural networks, which call for novel building blocks, provide an alternative physical embodiment of neural networks featuring a lower static power and a better emulation of neural functionalities. Here, we develop neuro-transistors by integrating dynamic pseudo-memcapacitors as the gates of transistors to produce electronic analogs of the soma and axon of a neuron, with “leaky integrate-and-fire” dynamics augmented by a signal gain on the output. Paired with non-volatile pseudo-memcapacitive synapses, a Hebbian-like learning mechanism is implemented in a capacitive switching network, leading to the observed associative learning. A prototypical fully integrated capacitive neural network is built and used to classify inputs of signals.

SPEC CPU2017: Next-Generation Compute Benchmark

Abstract: Description of the new features of the SPEC CPU2017 industry standard benchmark and its metric calculations.

Defective Hard Carbon Anode for Na-Ion Batteries

Abstract: Hard carbon as an anode is critical for the near-future commercialization of Na-ion batteries. However, where Na ions are located at different states of charge with respect to the local structures of hard carbon remains a topic that is under debate. Recently, some groups, including ours, have suggested a structure–property correlation that assigns the slope capacity in galvanostatic charge/discharge curves to the binding of Na ions to structural defects of hard carbon. To test this correlation, herein, we prepared a highly defective hard carbon by microwaving a carbon that was obtained by pyrolysis of cellulose at 650 °C. After this microwave treatment for just 6 s, the reversible capacity of the hard carbon increased from 204 to 308 mAh/g, which is significantly higher than that of hard carbon annealed at 1100 °C for 7 h (274 mAh/g). The microwave treatment not only is energy-efficient but also retains a high extent of the structural vacancies in hard carbon, as demonstrated by neutron total scattering a...

SuRF: Practical Range Query Filtering with Fast Succinct Tries

Abstract: We present the Succinct Range Filter (SuRF), a fast and compact data structure for approximate membership tests. Unlike traditional Bloom filters, SuRF supports both single-key lookups and common range queries: open-range queries, closed-range queries, and range counts. SuRF is based on a new data structure called the Fast Succinct Trie (FST) that matches the point and range query performance of state-of-the-art order-preserving indexes, while consuming only 10 bits per trie node. The false positive rates in SuRF for both point and range queries are tunable to satisfy different application needs. We evaluate SuRF in RocksDB as a replacement for its Bloom filters to reduce I/O by filtering requests before they access on-disk data structures. Our experiments on a 100 GB dataset show that replacing RocksDB's Bloom filters with SuRFs speeds up open-seek (without upper-bound) and closed-seek (with upper-bound) queries by up to 1.5× and 5× with a modest cost on the worst-case (all-missing) point query throughput due to slightly higher false positive rate.

Enhancing Catalytic Activity of MoS2 Basal Plane S-Vacancy by Co Cluster Addition

Abstract: The basal plane of molybdenum disulfide (MoS2) was recently activated for hydrogen evolution reaction (HER) by creating sulfur (S) vacancies (MoS2–x). However, the HER activity of those S-vacancies depends on the concentration of S-vacancies, imposing a dilemma for either improving activity per site or increasing overall active site density. Herein, we use density functional theory (DFT) calculations and experiments to show that the HER activities of MoS2–x are greatly enhanced by adding cobalt (Co) clusters on the basal plane. Our DFT results show that the highest HER activity is achieved when the Co clusters are anchored on the S-vacancies with the interface of Co–Mo as the preferred active site. Our experiments confirm that the addition of Co enhances the activity per unit active site and increases the electrochemical active surface area. These results demonstrate the basal plane activity of MoS2–x can be enhanced by decorating S-vacancies with transition-metal clusters.

Microprocessor Optimizations for the Internet of Things: A Survey

Abstract: The Internet of Things (IoT) refers to a pervasive presence of interconnected and uniquely identifiable physical devices. These devices’ goal is to gather data and drive actions in order to improve productivity, and ultimately reduce or eliminate reliance on human intervention for data acquisition, interpretation, and use. The proliferation of these connected low-power devices will result in a data explosion that will significantly increase data transmission costs with respect to energy consumption and latency. Edge computing reduces these costs by performing computations at the edge nodes, prior to data transmission, to interpret and/or utilize the data. While much research has focused on the IoT’s connected nature and communication challenges, the challenges of IoT embedded computing with respect to device microprocessors has received much less attention. This paper explores IoT applications’ execution characteristics from a microarchitectural perspective and the microarchitectural characteristics that will enable efficient and effective edge computing. To tractably represent a wide variety of next-generation IoT applications, we present a broad IoT application classification methodology based on application functions, to enable quicker workload characterizations for IoT microprocessors. We then survey and discuss potential microarchitectural optimizations and computing paradigms that will enable the design of right-provisioned microprocessors that are efficient, configurable, extensible, and scalable. This paper provides a foundation for the analysis and design of a diverse set of microprocessor architectures for next-generation IoT devices.

Design Space Exploration of Microring Resonators in Silicon Photonic Interconnects: Impact of the Ring Curvature

Abstract: A detailed analysis of fundamental tradeoffs between ring radius and coupling gap size is presented to draw realistic borders of the possible design space for microring resonators (MRRs). The coupling coefficient for the ring-waveguide structure is estimated based on an integration of the nonuniform gap between the ring and the waveguide. Combined with the supermode analysis of two coupled waveguides, this approach is further expanded into a closed-form equation that describes the coupling strength. This equation permits to evaluate how the distance separating a waveguide from a ring resonator, and the ring radius, affect coupling. The effect of ring radius on the bending loss of the ring is furthermore modeled based on the measurements for silicon MRRs with different radii. These compact models for coupling and loss are subsequently used to derive the main optical properties of MRRs, such as 3-dB optical bandwidth, extinction ratio of resonance, and insertion loss, hence identifying the design space. Our results indicate that the design space for add-drop filters in a wavelength division multiplexed link is currently limited to 5–10 $\mu$ m in radius and gap sizes ranging from 120 to 210 nm. The good agreement between the results from the proposed compact model for coupling and the numerical FDTD and experimental measurements indicate the application of our approach in realizing fast and efficient design space exploration of MRRs in silicon photonic interconnects.

Towards Scalable and Ubiquitous Millimeter-Wave Wireless Networks

Abstract: Millimeter-wave (mmWave) technology is emerging as the most promising solution to meet the multi-fold demand increase for mobile data. Very short wavelength, high directionality, together with sensitivity to rampant blockages and mobility, however, render state-of-the-art mmWave technologies unsuitable for ubiquitous wireless coverage. In this work, we design and implement UbiG - a mmWave wireless access network - that can deliver ubiquitous gigabits per second wireless access consistently to the commercial-off-the-shelf IEEE 802.11ad devices. UbiG has two key design components: (1) a fast probing based beam alignment algorithm that can identify the best beam consistently with guaranteed latency in a mmWave link, and the algorithm scales well even with a very large number of beams; and (2) an infrastructure-side predictive ranking based fast access point switching algorithm to ensure seamless gigabits per second connectivity under mobility and blockage in a dense mmWave deployment. Our IEEE 802.11ad testbed experiments show that UbiG performs close to an "Oracle" solution that instantaneously knows the best beam and access point for gigabits per second data transmission to users.

Influence Maximization in Online Social Networks

Abstract: Starting with the earliest studies showing that the spread of new trends, information, and innovations is closely related to the social influence exerted on people by their social networks, the research on social influence theory took off, providing remarkable evidence on social influence induced viral phenomena. Fueled by the extreme popularity of online social networks and social media, computational social influence has emerged as a subfield of data mining whose goal is to analyze and optimize social influence using computational frameworks such as algorithm design and theoretical modeling. One of the fundamental problems in this field is the problem of influence maximization, primarily motivated by the application of viral marketing. The objective is to identify a small set of users in a social network who, when convinced to adopt a product, shall influence others in the network in a manner that leads to a large number of adoptions. In this tutorial, we extensively survey the research on social influence propagation and maximization, with a focus on the recent algorithmic and theoretical advances. To this end, we provide detailed reviews of the latest research effort devoted to (i) improving the efficiency and scalability of the influence maximization algorithms; (ii) context-aware modeling of the influence maximization problem to better capture real-world marketing scenarios; (iii) modeling and learning of real-world social influence; (iv) bridging the gap between social advertising and viral marketing.

MAPLE 2.3.0: an improved system for evaluating the functionomes of genomes and metagenomes.

Abstract: MAPLE is an automated system for inferring the potential comprehensive functions harbored by genomes and metagenomes. To reduce runtime in MAPLE analyzing the massive amino acid datasets of over 1 million sequences, we improved it by adapting the KEGG automatic annotation server to use GHOSTX and verified no substantial difference in the MAPLE results between the original and new implementations.

Continuous-Time Flows for Efficient Inference and Density Estimation

Abstract: Two fundamental problems in unsupervised learning are efficient inference for latent-variable models and robust density estimation based on large amounts of unlabeled data Algorithms for the two tasks, such as normalizing flows and generative adversarial networks (GANs), are often developed independently In this paper, we propose the concept of {\em continuous-time flows} (CTFs), a family of diffusion-based methods that are able to asymptotically approach a target distribution Distinct from normalizing flows and GANs, CTFs can be adopted to achieve the above two goals in one framework, with theoretical guarantees Our framework includes distilling knowledge from a CTF for efficient inference, and learning an explicit energy-based distribution with CTFs for density estimation Both tasks rely on a new technique for distribution matching within amortized learning Experiments on various tasks demonstrate promising performance of the proposed CTF framework, compared to related techniques

Separation of current density and electric field domains caused by nonlinear electronic instabilities

Abstract: In 1963 Ridley postulated that under certain bias conditions circuit elements exhibiting a current- or voltage-controlled negative differential resistance will separate into coexisting domains with different current densities or electric fields, respectively, in a process similar to spinodal decomposition of a homogeneous liquid or disproportionation of a metastable chemical compound. The ensuing debate, however, failed to agree on the existence or causes of such electronic decomposition. Using thermal and chemical spectro-microscopy, we directly imaged signatures of current-density and electric-field domains in several metal oxides. The concept of local activity successfully predicts initiation and occurrence of spontaneous electronic decomposition, accompanied by a reduction in internal energy, despite unchanged power input and heat output. This reveals a thermodynamic constraint required to properly model nonlinear circuit elements. Our results explain the electroforming process that initiates information storage via resistance switching in metal oxides and has significant implications for improving neuromorphic computing based on nonlinear dynamical devices.

Error-Free Operation in a Hybrid-Silicon Quantum Dot Comb Laser

Abstract: We investigate amplitude noise in a wafer-bonded quantum dot laser on silicon. Error-free operation at room temperature and under continuous current injection in the three highest-power channels is observed with a signal-to-noise ratio of 11.5 dB or larger. These devices are attractive candidates as an optical engine for interconnects in next-generation data centers and exascale computers.

In-Memory Computing with Memristor Arrays

Abstract: Memristors with tunable non-volatile resistance states offer the potential for in-memory computing that mitigates the von-Neumann bottleneck. We build a large scale memristor array by integrating a transistor array with Ta/HfO2 memristors that have stable multilevel resistance states and linear IV characteristic. With off-chip peripheral driving circuits, the memristor chip is capable of high-precision analog computing and online learning. We demonstrate a weight-update scheme that provides linear and symmetric potentiation and depression with no more than two pulses for each cell. We train the array as a single-layer fully-connected feedforward neural network for the WDBC data base and achieve 98% classification accuracy. We further partition the array into a two-layer network, which achieves 91.71% classification accuracy for MNIST database experimentally. The system demonstrates high defect tolerance and excellent speed-energy efficiency.

Newton: Gravitating Towards the Physical Limits of Crossbar Acceleration

Abstract: Many recent works take advantage of highly parallel analog in-situ computation in memristor crossbars to accelerate the many vector-matrix multiplication operations in deep neural networks (DNNs). However, these in-situ accelerators have two significant shortcomings: The ADCs account for a large fraction of chip power and area, and these accelerators adopt a homogeneous design in which every resource is provisioned for the worst case. By addressing both problems, the new architecture, called Newton, moves closer to achieving optimal energy per neuron for crossbar accelerators. We introduce new techniques that apply at different levels of the tile hierarchy, some leveraging heterogeneity and others relying on divide-and-conquer numeric algorithms to reduce computations and ADC pressure. Finally, we place constraints on how a workload is mapped to tiles, thus helping reduce resource-provisioning in tiles. For many convolutional-neural-network (CNN) dataflows and structures, Newton achieves a 77-percent decrease in power, 51-percent improvement in energy-efficiency, and 2.1× higher throughput/area, relative to the state-of-the-art In-Situ Analog Arithmetic in Crossbars (ISAAC) accelerator.

The many faces of data-centric workflow optimization: a survey

Abstract: Workflow technology is rapidly evolving and, rather than being limited to modeling the control flow in business processes, is becoming a key mechanism to perform advanced data management, such as big data analytics. This survey focuses on data-centric workflows (or workflows for data analytics or data flows), where a key aspect is data passing through and getting manipulated by a sequence of steps. The large volume and variety of data, the complexity of operations performed, and the long time such workflows take to compute give rise to the need for optimization. In general, data-centric workflow optimization is a technology in evolution. This survey focuses on techniques applicable to workflows comprising arbitrary types of data manipulation steps and semantic inter-dependencies between such steps. Further, it serves a twofold purpose: firstly, to present the main dimensions of the relevant optimization problems and the types of optimizations that occur before flow execution and secondly, to provide a concise overview of the existing approaches with a view to highlighting key observations and areas deserving more attention from the community.

High performance low-temperature solid oxide fuel cells with atomic layer deposited-yttria stabilized zirconia embedded thin film electrolyte

Abstract: We report the design and the fabrication of a sandwich-structured thin-film composite electrolyte for low-temperature solid oxide fuel cells supported by a nanoporous anodized aluminum oxide (AAO) substrate We adopt an extremely thin (25 nm) atomic-layer-deposited (ALD) yttria-stabilized zirconia (YSZ) layer embedded in highly ionic-conductive sputtered samaria-doped ceria (SDC) electrolytes (360 nm thick) and systematically vary the position of the YSZ layer to investigate the optimal design of the composite electrolyte The cell with a sputtered SDC (180 nm)/ALD YSZ (25 nm)/sputtered SDC (180 nm) sandwich electrolyte shows a high maximum power density of 562 mW cm−2 at 450 °C, which is the highest performance at this temperature range for AAO-supported cells to date

On-Chip Hybrid Silicon Quantum Dot Comb Laser with 14 Error-Free Channels

Abstract: We present a quantum dot-based hybrid silicon comb laser using wafer bonding. Multimode interferometer-based on-chip mirrors and grating couplers are integrated on the silicon for wafer-level testing. We show error-free operation in 14 channels.

Hardware-Software Co-Design for an Analog-Digital Accelerator for Machine Learning

Abstract: The increasing deployment of machine learning at the core and at the edge for applications such as video and image recognition has resulted in a number of special purpose accelerators in this domain. However, these accelerators do not have full end-to-end software stacks for application development, resulting in hard-to-develop, proprietary, and suboptimal application programming and executables. In this paper, we describe software stack for a memristor-based hybrid (analog-digital)accelerator. The software stack consists of an ONNX converter, an application optimizer, a compiler, a driver, and emulators. The ONNX converter helps leveraging interoperable neural network models developed on frameworks that support ONNX, such as CNTK, Caffe2, Tensorflow, etc. The application optimization layer adapts these interoperable models to the underlying hardware. The compiler generates executable ISA code that the underlying accelerator can run. Finally, the emulator enables software execution without actual hardware which enables hardware design space exploration and testing. By building a software stack, we have made hybrid memristor-based ML accelerators more accessible to software developers, enabled a generation of better-performing executables, and created an environment that can be leveraged by a multitude of existing neural network models developed using other frameworks to target these accelerators.