Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Reconfigurable Computing”

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.

At least 559 records · Page 31

Towards Low-Overhead Resilience for Data Parallel Deep Learning

Data parallel techniques have been widely adopted both in academia and industry as a tool to enable scalable training of deep learning models. At scale, DL training jobs can fail due to software or hardware bugs, may need to be preempted or terminated due to unexpected events, or may perform suboptimally because they were misconfigured. Under such circumstances, there is a need to recover and/or reconfigure data-parallel DL training jobs on-the-fly, while minimizing the impact on the accuracy of the DNN model and the runtime overhead. In this regard, state-of-art techniques adopted by the HPC community mostly rely on checkpoint-restart, which inevitably leads to loss of progress, thus increasing the runtime overhead. In this paper we explore alternative techniques that exploit the properties of modern deep learning frameworks (overlapping of gradient averaging and weight updates with local gradient computations through pipeline parallelism) to reduce the overhead of resilience/elasticity. To this end we introduce a failure simulation framework and two resilience strategies (immediate mini-batch rollback and lossy forward recovery), which we study compared with checkpoint-restart approaches in a variety of settings in order to understand the trade-offs between the accuracy loss of the DNN model and the runtime overhead.

data-parallel training↗

Design and Stability Analysis of Control System in Multiport Autonomous Reconfigurable Solar Power Plants (MARS)

The Multiport Autonomous Reconfigurable Solar Power Plant (MARS) is an integrated photovoltaic (PV) power generation and energy storage system (ESS), that is designed to connect to both alternating current (AC) transmission grids and high-voltage direct current (HVDC) links. It is a three-phase plant consisting of numerous components with a complex hardware and hierarchical control architecture. This paper presents an approach to decouple the multivariable system of MARS using a recursive reduced-order and boundary layer system methodology. This approach enables efficient computation of the control parameters for the Ll, L2, and L3 controllers. To validate the effectiveness of the proposed control strategy, cyclic tests in accordance with pre-defined performance criteria using controller Hardware-in-the-Loop (cHIL) experiments are conducted. The results demonstrate that the MARS system operates consistently under steady-state conditions. Furthermore, the dynamic response of the MARS system to various grid events is analyzed, underlining the resilience of MARS in presence of faults or loss of generation within the connected WECC system.

Xia, Qian↗

Assurance Technology Challenges of Advanced Space Systems

The initiative to explore space and extend a human presence across our solar system to revisit the moon and Mars post enormous technological challenges to the nation's space agency and aerospace industry. Key areas of technology development needs to enable the endeavor include advanced materials, structures and mechanisms; micro/nano sensors and detectors; power generation, storage and management; advanced thermal and cryogenic control; guidance, navigation and control; command and data handling; advanced propulsion; advanced communication; on-board processing; advanced information technology systems; modular and reconfigurable systems; precision formation flying; solar sails; distributed observing systems; space robotics; and etc. Quality assurance concerns such as functional performance, structural integrity, radiation tolerance, health monitoring, diagnosis, maintenance, calibration, and initialization can affect the performance of systems and subsystems. It is thus imperative to employ innovative nondestructive evaluation methodologies to ensure quality and integrity of advanced space systems. Advancements in integrated multi-functional sensor systems, autonomous inspection approaches, distributed embedded sensors, roaming inspectors, and shape adaptive sensors are sought. Concepts in computational models for signal processing and data interpretation to establish quantitative characterization and event determination are also of interest. Prospective evaluation technologies include ultrasonics, laser ultrasonics, optics and fiber optics, shearography, video optics and metrology, thermography, electromagnetics, acoustic emission, x-ray, data management, biomimetics, and nano-scale sensing approaches for structural health monitoring.

Chern, E. James↗

A HOL theory for voting

Central to fault-tolerant computing is redundancy management, and common to proofs of fault-tolerance is a maximum fault assumption. Typically a maximum fault assumption is rather restrictive. Usually, this is necessary to avoid assumptions about the behavior of faulty channels. A maximum fault assumption is useful because it allows reasoning about fault tolerance in the presence of arbitrarily malicious fault behavior. However, analysis of the architecture may establish certain scenarios in which the assumption may be weakened. Proofs comparing majority and plurality and proofs of simple reconfiguration strategies are presented in viewgraph form.

Miner, Paul S.↗

An organic electrochemical transistor for multi-modal sensing, memory and processing

By integrating sensing, memory and processing functionalities, biological nervous systems are energy and area efficient. Emulating such capabilities in artificial systems is, however, challenging and is limited by the device heterogeneity of sensing and processing cores. Here we report an organic electrochemical transistor capable of sensing, memory and processing. The device has a vertical traverse architecture and a crystalline–amorphous channel that can be selectively doped by ions to enable two reconfigurable modes: a volatile receptor and a non-volatile synapse. As a volatile receptor, the device is capable of multi-modal sensing and is responsive to stimuli such as ions and light. As a non-volatile synapse, it is capable of 10-bit analogue states, low switching stochasticity and good state retention. We also show that the homogeneous integration of the devices could provide functions such as conditioned reflexes and could be used for real-time cardiac disease diagnoses via reservoir computing.

97 MATHEMATICS AND COMPUTING↗

Time Varying Compensator Design for Reconfigurable Structures Using Non-Collocated Feedback

Analysis and synthesis tools are developed to improved the dynamic performance of reconfigurable nonminimum phase, nonstrictly positive real-time variant systems. A novel Spline Varying Optimal (SVO) controller is developed for the kinematic nonlinear system. There are several advantages to using the SVO controller, in which the spline function approximates the system model, observer, and controller gain. They are: The spline function approximation is simply connected, thus the SVO controller is more continuous than traditional gain scheduled controllers when implemented on a time varying plant; ft is easier for real-time implementations in storage and computational effort; where system identification is required, the spline function requires fewer experiments, namely four experiments; and initial startup estimator transients are eliminated. The SVO compensator was evaluated on a high fidelity simulation of the Shuttle Remote Manipulator System. The SVO controller demonstrated significant improvement over the present arm performance: (1) Damping level was improved by a factor of 3; and (2) Peak joint torque was reduced by a factor of 2 following Shuttle thruster firings.

Scott, Michael A.↗

Exascale Virtualized and Programmable Distributed Cyber Resource Control: Final Scientific Technical Report

Extreme-science drives the need for distributed exascale processing and communications that are carefully, yet flexibly, managed. Exponential growth of data for scientific simulations, experimental data, collaborative data analyses, remote visualization and GRID computing requirements of scientists in fields as diverse as high energy physics, climate change, genomics, fusion, synchrotron radiation, material science, medicine, and other scientific disciplines cannot be accommodated by simply applying existing transport protocols to faster pipes. Further, scientific challenges today demand diverse research teams, heightening the need for and increasing the complexity of collaboration. To address these issues within the network layer and physical layer, we have performed a number of research activities surrounding effective allocation and management of elastic optical network (EON) resources, particularly focusing on FlexGrid transponders. FlexGrid transponders support the opportunity to build Layer-1 connections at a wide range of bandwidths and to reconfigure them rapidly. The new flexibility supports complex new ways of using the physical layer that must be carefully managed and hidden from the scientist end-users. FlexGrid networks utilize flexible (or elastic) spectral bandwidths for each data link without using fixed wavelength grids. The flexibility in spectrum allocation brings many appealing features to network operations. Current networks are designed for the worst case impairments in transmission performance and the assigned spectrum is over-provisioned. In contrast, the FlexGrid networks can operate with the highest spectral efficiency and minimum bandwidth for the given traffic demand while meeting the minimum quality of transmission (QoT) requirement. Two primary focuses of our research are: (1) resource and spectrum allocation (RSA) for IP traffic over EONs, and (2) RSA for cross-domain optical networks. Previous work concentrates primarily on large file transfers within a single domain. Adding support for IP traffic changes the nature of the RSA problem: instead of choosing to accept or deny each request for network support, IP traffic is inherently elastic and thus lends itself to a bandwidth maximization formulation. We developed a number of algorithms that could be easily deployed within existing and new FlexGrid networks, leading to networks that better support scientific collaboration. Cross-domain RSA research is essential to support large-scale FlexGrid networks, since configuration information is generally not shared or coordinated across domains. The results presented here are in their early stages. They are technically feasible and practical, but still require coordination among organizations and equipment owners and a higher-layer framework for managing network requests.

42 ENGINEERING↗

Multimodal user input to supervisory control systems - Voice-augmented keyboard

The use of a voice-augmented keyboard input modality is evaluated in a supervisory control application. An implementation of voice recognition technology in supervisory control is proposed: voice is used to request display pages, while the keyboard is used to input system reconfiguration commands. Twenty participants controlled GT-MSOCC, a high-fidelity simulation of the operator interface to a NASA ground control system, via a workstation equipped with either a single keyboard or a voice-augmented keyboard. Experimental results showed that in all cases where significant performance differences occurred, performance with the voice-augmented keyboard modality was inferior to and had greater variance than the keyboard-only modality. These results suggest that current moderately priced voice recognition systems are an inappropriate human-computer interaction technology in supervisory control systems.

Mitchell, Christine M.↗

The use of automatic programming techniques for fault tolerant computing systems

It is conjectured that the production of software for ultra-reliable computing systems such as required by Space Station, aircraft, nuclear power plants and the like will require a high degree of automation as well as fault tolerance. In this paper, the relationship between automatic programming techniques and fault tolerant computing systems is explored. Initial efforts in the automatic synthesis of code from assertions to be used for error detection as well as the automatic generation of assertions and test cases from abstract data type specifications is outlined. Speculation on the ability to generate truly diverse designs capable of recovery from errors by exploring alternate paths in the program synthesis tree is discussed. Some initial thoughts on the use of knowledge based systems for the global detection of abnormal behavior using expectations and the goal-directed reconfiguration of resources to meet critical mission objectives are given. One of the sources of information for these systems would be the knowledge captured during the automatic programming process.

Wild, C.↗

SpaceCube 2.0: An Advanced Hybrid Onboard Data Processor

The SpaceCube 2.0 is a compact, high performance, low-power onboard processing system that takes advantage of cutting-edge hybrid (CPU/FPGA/DSP) processing elements. The SpaceCube 2.0 design concept includes two commercial Virtex-5 field-programmable gate array (FPGA) parts protected by gradiation hardened by software" technology, and possesses exceptional size, weight, and power characteristics [5x5x7 in., 3.5 lb (approximately equal to 12.7 x 12.7 x 17.8 cm, 1.6 kg) 5-25 W, depending on the application fs required clock rate]. The two Virtex-5 FPGA parts are implemented in a unique back-toback configuration to maximize data transfer and computing performance. Draft computing power specifications for the SpaceCube 2.0 unit include four PowerPC 440s (1100 DMIPS each), 500+ DSP48Es (2x580 GMACS), 100+ LVDS high-speed serial I/Os (1.25 Gbps each), and 2x190 GFLOPS single-precision (65 GFLOPS double-precision) floating point performance. The SpaceCube 2.0 includes PROM memory for CPU boot, health and safety, and basic command and telemetry functionality; RAM memory for program execution; and FLASH/EEPROM memory to store algorithms and application code for the CPU, FPGA, and DSP processing elements. Program execution can be reconfigured in real time and algorithms can be updated, modified, and/or replaced at any point during the mission. Gigabit Ethernet, Spacewire, SATA and highspeed LVDS serial/parallel I/O channels are available for instrument/sensor data ingest, and mission-unique instrument interfaces can be accommodated using a compact PCI (cPCI) expansion card interface. The SpaceCube 2.0 can be utilized in NASA Earth Science, Helio/Astrophysics and Exploration missions, and Department of Defense satellites for onboard data processing. It can also be used in commercial communication and mapping satellites.

Lin, Michael↗

Fault-Tolerant Software-Defined Radio on Manycore

Software-defined radio (SDR) platforms generally rely on field-programmable gate arrays (FPGAs) and digital signal processors (DSPs), but such architectures require significant software development. In addition, application demands for radiation mitigation and fault tolerance exacerbate programming challenges. MaXentric Technologies, LLC, has developed a manycore-based SDR technology that provides 100 times the throughput of conventional radiationhardened general purpose processors. Manycore systems (30-100 cores and beyond) have the potential to provide high processing performance at error rates that are equivalent to current space-deployed uniprocessor systems. MaXentric's innovation is a highly flexible radio, providing over-the-air reconfiguration; adaptability; and uninterrupted, real-time, multimode operation. The technology is also compliant with NASA's Space Telecommunications Radio System (STRS) architecture. In addition to its many uses within NASA communications, the SDR can also serve as a highly programmable research-stage prototyping device for new waveforms and other communications technologies. It can also support noncommunication codes on its multicore processor, collocated with the communications workload-reducing the size, weight, and power of the overall system by aggregating processing jobs to a single board computer.

Ricketts, Scott↗

Development of a wind turbine model and simulation platform using an acausal approach: Multiphysics modeling, validation, and control

This article presents the development of the Control-oriented, Reconfigurable, and Acausal Floating Turbine Simulator (CRAFTS). CRAFTS has a modular, hierarchical model architecture that enables rapid and accurate simulation of wind turbines. The architecture facilitates the incorporation of model variants, and its system reconfiguration features help simulate multiple design variants. CRAFTS also supports the integration of models developed on a causality-free platform (e.g., Modelica®) with existing causal models. This article focuses on the validation of a land-based wind turbine models against OpenFAST, an industry-standard platform, for several test cases. Closed-loop scenarios are also tested using the standard ROSCO controller and compared against OpenFAST. In addition, a nonlinear controller developed in our prior work is also evaluated. The test cases demonstrate the user-friendly and computationally efficient capabilities of CRAFTS to facilitate control co-design, assist in incorporating multiphysics models, be adaptable to design variants, and allow for rapid simulations to validate models and evaluate controllers.

17 WIND ENERGY↗

Hierarchical Materials from High Information Content Macromolecular Building Blocks: Construction, Dynamic Interventions, and Prediction

Hierarchical materials that exhibit order over multiple length scales are ubiquitous in nature. Because hierarchy gives rise to unique properties and functions, many have sought inspiration from nature when designing and fabricating hierarchical matter. More and more, however, nature’s own high-information content building blocks, proteins, peptides, and peptidomimetics, are being coopted to build hierarchy because the information that determines structure, function, and interfacial interactions can be readily encoded in these versatile macromolecules. Here, we take stock of recent progress in the rational design and characterization of hierarchical materials produced from high-information content blocks with a focus on stimuli-responsive and “smart” architectures. We also review advances in the use of computational simulations and data-driven predictions to shed light on how the side chain chemistry and conformational flexibility of macromolecular blocks drive the emergence of order and the acquisition of hierarchy and also on how ionic, solvent, and surface effects influence the outcomes of assembly. Furthermore, continued progress in the above areas will ultimately usher in an era where an understanding of designed interactions, surface effects, and solution conditions can be harnessed to achieve predictive materials synthesis across scale and drive emergent phenomena in the self-assembly and reconfiguration of high-information content building blocks.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Lifting and Dropping VMs to Dynamically Transition Between Time- and Space-sharing for Large-Scale HPC Systems

As HPC environments increasingly integrate with edge based systems, system architectures will need to handle a broader class of workloads and scheduling requirements. One result of this shift will be the need to simultaneously support bulk-synchronous parallel (BSP) and on-demand service based applications on the same infrastructure. This in turn will require that future resource management approaches utilize both space-shared as well as time-shared resource scheduling strategies. In this work we introduce the concept of "VM-lifting'' (and its inverse "VM-Dropping'') which allows dynamically switching an HPC workload between space-shared and time-shared scheduling regimes. Our work targets co-kernel based HPC system software environments, in which multiple specialized OS kernels execute natively on dedicated physical resource partitions inside a single compute node. With VM-lifting, a native co-kernel can be migrated at runtime to and from locally hosted Virtual Machine Environments due to changing scheduling requirements of the node. This allows an HPC node to be dynamically (re-)configured as either a time-shared Infrastructure-as-a-Service (IaaS) resource or a dedicated space shared resource based on the current workload demands. We have implemented this approach in the context of the Hobbes Exascale System Software stack and have demonstrated that a node can be reconfigured with minimal impact on the running applications.

Gordon, Nick↗

Programmer-guided reliability for extreme-scale applications

We present “programmer-guided reliability” (PGR) as a systematic conceptual approach to address the expected rise in soft errors in coming extreme-scale systems at the application level. The approach involves instrumentation of the application with code to detect data corruption errors. The location and nature of these error detectors are at the discretion of the programmer, who uses their knowledge and experience with the problem domain, the application, the solution algorithms, etc., to determine the most vulnerable areas of the code and the most appropriate ways to detect data corruption. To illustrate the approach, we provide examples of error detectors from four different benchmark-scale applications. We also describe a simple control framework that allows for runtime configuration of the error detectors without recompilation of the application, as well as dynamic reconfiguration during the execution of the application. Lastly, we discuss a number of future directions building on the basic PGR approach, including the incorporation of some general error detectors into the programming environment in order to make them more easily usable by the programmer.

97 MATHEMATICS AND COMPUTING↗

Superconducting disordered neural networks for neuromorphic processing with fluxons

In superconductors, magnetic fields are quantized into discrete fluxons (flux quanta Φ 0 ), made of microscopic circulating supercurrents. We introduce a multiterminal synapse network comprising a disordered array of superconducting loops with Josephson junctions. The loops can trap fluxons defining memory, while the junctions allow their movement between loops. Dynamics of fluxons through such a disordered system through a complex reconfigurable energy landscape represents brain-like spiking information flow. In this work, we experimentally demonstrate a three-loop network using YBa 2 Cu 3 O 7 –δ -based superconducting loops and Josephson junctions, which exhibit stable memory configurations of trapped flux in loops that determine the rate of flow of fluxons through synaptic connections. The memory states are, in turn, affected by the applied input signals but can also be externally configured electrically through control current/feedback terminals. These results establish a previously unexplored, biologically similar architectural approach to neuromorphic computing that is scalable while dissipating energy of atto Joules/spike.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Advanced information processing system: Authentication protocols for network communication

In safety critical I/O and intercomputer communication networks, reliable message transmission is an important concern. Difficulties of communication and fault identification in networks arise primarily because the sender of a transmission cannot be identified with certainty, an intermediate node can corrupt a message without certainty of detection, and a babbling node cannot be identified and silenced without lengthy diagnosis and reconfiguration . Authentication protocols use digital signature techniques to verify the authenticity of messages with high probability. Such protocols appear to provide an efficient solution to many of these problems. The objective of this program is to develop, demonstrate, and evaluate intercomputer communication architectures which employ authentication. As a context for the evaluation, the authentication protocol-based communication concept was demonstrated under this program by hosting a real-time flight critical guidance, navigation and control algorithm on a distributed, heterogeneous, mixed redundancy system of workstations and embedded fault-tolerant computers.

Harper, Richard E.↗

A Reconfiguration Scheme for Accommodating Actuator Failures in Multi-Input, Multi-Output Flight Control Systems

A multi-input, multi-output reconfigurable flight control system design utilizing a robust controller and an adaptive filter is presented. The robust control design consists of a reduced-order, linear dynamic inversion controller with an outer-loop compensation matrix derived from Quantitative Feedback Theory (QFT). A principle feature of the scheme is placement of the adaptive filter in series with the QFT compensator thus exploiting the inherent robustness of the nominal flight control system in the presence of plant uncertainties. An example of the scheme is presented in a pilot-in-the-loop computer simulation using a simplified model of the lateral-directional dynamics of the NASA F18 High Angle of Attack Research Vehicle (HARV) that included nonlinear anti-wind up logic and actuator limitations. Prediction of handling qualities and pilot-induced oscillation tendencies in the presence of these nonlinearities is included in the example.

Siwakosit, W.↗