Search NASA⌕ Search

SEARCH · Search NASA

Results for “TRAINING PLATFORMS”

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 235 records · Page 13
All sourcesNASA NTRSDOE OSTI
NASA NTRSDate not supplied

Full Lunar Surface Visualization and Simulation Platform

This paper reports the results of the innovative work completed to facilitate the creation of a lunar surface visualization and simulation platform for the entire surface of the moon as a design, test, training, verification, and mission support environment. The mathematical approach to the visualization of the data sets utilizes fractal mathematics to generate not only the highest possible detail that the Lunar Reconnaissance Orbiter (LRO) data can provide, but it also additionally supplies an estimation of surface reflectivity and lighting based upon location on the lunar surface. This level of detail and accuracy is required to conduct accurate integrated crew / vehicle risk assessment throughout the design-to-disposal life cycle of vehicles and their supporting procedures and mission priorities. This process starts with safety assessments of integrated hardware / software / crew/ and support environments. Unlike the Apollo Program, the Artemis Program is not risk tolerant.

Simulation↗
DOE OSTI2022

Secure Collaborative Environment for Seamless Sharing of Scientific Knowledge

In a secure collaborative environment, tera-bytes of data generated from powerful scientific instruments are used to train secure machine learning (ML) models on exascale computing systems, which are then securely shared with internal or external collaborators as cloud-based services. Devising such a secure platform is necessary for seamless scientific knowledge sharing without compromising individual, or institute-level, intellectual property and privacy details. By enabling new computing opportunities with sensitive data, we envision a secure collaborative environment that will play a significant role in accelerating scientific discovery. Several recent technological advancements have made it possible to realize these capabilities. In this paper, we present our efforts at ORNL toward developing a secure computation platform. We present a use case where scientific data generated from complex instruments, like those at the Spallation Neutron Source (SNS), are used to train a differential privacy enabled deep learning (DL) network on Summit, which is then hosted as a secure multi-party computation (MPC) service on ORNL’s Compute and Data Environment for Science (CADES) cloud computing platform for third-party inference. In this feasibility study, we discuss the challenges involved, elaborate on leveraged technologies, analyze relevant performance results and present the future vision of our work to establish secure collaboration capabilities within and outside of ORNL.

Yoginath, Srikanth↗
DOE OSTI2024

Flexible silicon photonic architecture for accelerating distributed deep learning

The increasing size and complexity of deep learning (DL) models have led to the wide adoption of distributed training methods in datacenters (DCs) and high-performance computing (HPC) systems. However, communication among distributed computing units (CUs) has emerged as a major bottleneck in the training process. In this study, we propose Flex-SiPAC, a flexible silicon photonic accelerated compute cluster designed to accelerate multi-tenant distributed DL training workloads. Flex-SiPAC takes a co-design approach that combines a silicon photonic hardware platform with a tailored collective algorithm, optimized to leverage the unique physical properties of the architecture. The hardware platform integrates a novel wavelength-reconfigurable transceiver design and a micro-resonator-based wavelength-reconfigurable switch, enabling the system to achieve flexible bandwidth steering in the wavelength domain. The collective algorithm is designed to support reconfigurable topologies, enabling efficient all-reduce communications that are commonly used in DL training. The feasibility of the Flex-SiPAC architecture is demonstrated through two testbed experiments. First, an optical testbed experiment demonstrates the flexible routing of wavelengths by shuffling an array of input wavelengths using a custom-designed spatial-wavelength selective switch. Second, a four-GPU testbed running two DL workloads shows a 23% improvement in job completion time compared to a similarly sized leaf-spine topology. We further evaluate Flex-SiPAC using large-scale simulations, which show that Flex-SiPAC is able to reduce the communication time by 26% to 29% compared to state-of-the-art compute clusters under representative collective operations.

Wu, Zhenguo (ORCID:0000000322847985)↗
DOE OSTI2020

Creating multifunctional synthetic lichen platforms for sustainable biosynthesis of biofuel precursors

In this project, we were creating a sustainable platform for biofuel production, utilizing carbon-fixing autotrophs to supply oxygen and organic substrates to heterotrophic partners, which in turn produce carbon dioxide to feed the autotrophs. This symbiotic lichen community could lower the input cost, optimize metabolic exchanges and improve the generation of biofuel precursors through multi-omics driven genetic engineering. The cyanobacteria Synechococcus elongatus (S. elongatus) was used as the primary autotroph to provide oxygen and organic substrates, especially sucrose, to a co-culture system. The strain with overexpression of sucrose transporter cscB demonstrated a significant increase in sucrose production under salt stress as what we expected. We also implemented 13C metabolic flux analysis on the sucrose secreting strain S. elongatus cscB-NaCl. Next, transporters proteins like glutamate exporter mscCG from Corynebacterium glutamicum was overexpressed in S. elongatus to improve metabolite exchange. We provided sucrose and glutamate to filamentous fungi to enable their growth and production of biochemicals using substrates from the cyanobacterium. Two fungi ( Aspergillus nidulans and Aspergillus niger ) and two yeast strains ( Rhodotorula toruloides and Lipomyces starkeyi ) served as heterotrophs in this system. They were co-cultured with S.elongatus under different pH condition, and growth on different carbon source to understand the symbiotic system and optimize the parameters for production. The two fungi strained grew much better at pH 7, 8 and 9 with Yeast Nitrogen Base (YNB ) supplementation. Two yeast strains grew well on pH 7 and 8 with YNB supplementation. All four species achieved the highest biomass level when utilizing sucrose as the main carbon sources, which demonstrates a great pairing with S.elongatus . In addition to investigate the metabolite exchange and growth condition of the co-culture system, we incorporated an Aspergillus strain expressing three genes (PAND, BAPAT and HPDH). Expression of these three genes enabled production of 3-hydroxypropionic acid (3HP) with a titer of 3-6 g/L regardless of pH and other nutrient conditions. Moreover, we developed a computational model with different heterotrophic fungi symbiosing with S.elongatus to evaluate the exchange of hundreds of metabolites. The accuracy and sensitivity had been optimized by high-throughput phenotyping assays. This project demonstrated ways to enhance a synthetic lichen platform’s efficiency and scalability to create a robust system for sustainable bioproduct synthesis. The project has enabled us to explore the technological and commercial potential of sustainable lichen co-culture. The project has also trained the next generation of scientists to address sustainability, carbon fixation, and design solutions to contemporary global challenges. The lichen platform represented a novel approach to sustainably producing bioproducts with a reduced carbon footprint and lower costs.

09 BIOMASS FUELS↗
NASA NTRS2019

Next-Generation Optical Sensing Technologies for Exploring Ocean Worlds - NASA FluidCam, MiDAR, and NeMO-Net

We highlight three emerging NASA optical technologies that enhance our ability to remotely sense, analyze, and explore ocean worlds–FluidCam and fluid lensing, MiDAR, and NeMO-Net. Fluid lensing is the first remote sensing technology capable of imaging through ocean waves without distortions in 3D at sub-cm resolutions. Fluid lensing and the purpose-built FluidCam CubeSat instruments have been used to provide refraction-corrected 3D multispectral imagery of shallow marine systems from unmanned aerial vehicles (UAVs). Results from repeat 2013 and 2016 airborne fluid lensing campaigns over coral reefs in American Samoa present a promising new tool for monitoring fine-scale ecological dynamics in shallow aquatic systems tens of square kilometers in area. MiDAR is a recently-patented active multispectral remote sensing and optical communications instrument which evolved from FluidCam. MiDAR is being tested on UAVs and autonomous underwater vehicles (AUVs) to remotely sense living and non-living structures in light-limited and analog planetary science environments. MiDAR illuminates targets with high-intensity narrowband structured optical radiation to measure an object’s spectral reflectance while simultaneously transmitting data. MiDAR is capable of remotely sensing reflectance at fine spatial and temporal scales, with a signal-to-noise ratio 10-10(exp 3) times higher than passive airborne and spaceborne remote sensing systems, enabling high-framerate multispectral sensing across the ultraviolet, visible, and near-infrared spectrum. Preliminary results from a 2018 mission to Guam show encouraging applications of MiDAR to imaging coral from airborne and underwater platforms whilst transmitting data across the air-water interface. Finally, we share NeMO-Net, the Neural Multi-Modal Observation & Training Network for Global Coral Reef Assessment. NeMO-Net is a machine learning technology under development that exploits high-resolution data from FluidCam and MiDAR for augmentation of low-resolution airborne and satellite remote sensing. NeMO-Net is intended to harmonize the growing diversity of 2D and 3D remote sensing with in situ data into a single open-source platform for assessing shallow marine ecosystems globally using active learning for citizen-science based training. Preliminary results from four-class Q17 coral classification have an accuracy of 94.4%. Together, these maturing technologies present promising scalable, practical, and cost-efficient innovations that address current observational and technological challenges in optical sensing of marine systems.

Ved Chirayath↗
DOE OSTI2022

PowerNet: Multi-agent Deep Reinforcement Learning for Scalable Powergrid Control

This paper develops an efficient multi-agent deep reinforcement learning algorithm for cooperative controls in powergrids. Specifically, we consider the decentralized inverter-based secondary voltage control problem in distributed generators (DGs), which is first formulated as a cooperative multi-agent reinforcement learning (MARL) problem. We then propose a novel on-policy MARL algorithm, PowerNet, in which each agent (DG) learns a control policy based on (sub-)global reward but local states and encoded communication messages from its neighbors. Motivated by the fact that a local control from one agent has limited impact on agents distant from it, we exploit a novel spatial discount factor to reduce the effect from remote agents, to expedite the training process and improve scalability. Furthermore, a differentiable, learning-based communication protocol is employed to foster the collaborations among neighboring agents. In addition, to mitigate the effects of system uncertainty and random noise introduced during on-policy learning, we utilize an action smoothing factor to stabilize the policy execution. To facilitate training and evaluation, we develop PGSim, an efficient, high-fidelity powergrid simulation platform. Here, experimental results in two microgrid setups show that the developed PowerNet outperforms the conventional model-based control method, as well as several state-of-the-art MARL algorithms. The decentralized learning scheme and high sample efficiency also make it viable to large-scale power grids.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2016

High Energy Physics Exascale Requirements Review. An Office of Science review sponsored jointly by Advanced Scientific Computing Research and High Energy Physics, June 10-12, 2015, Bethesda, Maryland

The U.S. Department of Energy (DOE) Office of Science (SC) Offices of High Energy Physics (HEP) and Advanced Scientific Computing Research (ASCR) convened a programmatic Exascale Requirements Review on June 10–12, 2015, in Bethesda, Maryland. This report summarizes the findings, results, and recommendations derived from that meeting. The high-level findings and observations are as follows. Larger, more capable computing and data facilities are needed to support HEP science goals in all three frontiers: Energy, Intensity, and Cosmic. The expected scale of the demand at the 2025 timescale is at least two orders of magnitude — and in some cases greater — than that available currently. The growth rate of data produced by simulations is overwhelming the current ability of both facilities and researchers to store and analyze it. Additional resources and new techniques for data analysis are urgently needed. Data rates and volumes from experimental facilities are also straining the current HEP infrastructure in its ability to store and analyze large and complex data volumes. Appropriately configured leadership-class facilities can play a transformational role in enabling scientific discovery from these datasets. A close integration of high-performance computing (HPC) simulation and data analysis will greatly aid in interpreting the results of HEP experiments. Such an integration will minimize data movement and facilitate interdependent workflows. Long-range planning between HEP and ASCR will be required to meet HEP’s research needs. To best use ASCR HPC resources, the experimental HEP program needs (1) an established, long-term plan for access to ASCR computational and data resources, (2) the ability to map workflows to HPC resources, (3) the ability for ASCR facilities to accommodate workflows run by collaborations potentially comprising thousands of individual members, (4) to transition codes to the next-generation HPC platforms that will be available at ASCR facilities, (5) to build up and train a workforce capable of developing and using simulations and analysis to support HEP scientific research on next-generation systems.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

The Aerosol Model Benchmarking Repository: A toolkit for model intercomparison

The Aerosol Model Benchmarking Repository and Standards (AMBRS) project was initiated to provide tools and to establish community standards for benchmarking aerosol models. This report describes a set of open-source tools for building, running, and analyzing aerosol box model simulations in a standardized framework. The framework consists of three core components: AMBuilder, a CMake-based build system that compiles supported models consistently; AMBRS, a Python module that defines unified numerical experiments and executes them with aligned inputs; and PyParticle, an aerosol analysis package that standardizes output, computes diagnostics, and visualizes simulation results. Together, these tools enable reproducible intercomparison of aerosol schemes and support process-level evaluation of how model simplifications affect predictions of size distributions, cloud condensation nuclei activity, and other relevant properties relevant for the Earth-Energy system. Beyond its role in benchmarking, AMBRS provides a platform for studying aerosol processes across scales and can be used to generate training data for AI/ML applications in support of a broader hierarchical aerosol modeling strategy.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Applications and Techniques for Fast Machine Learning in Science

In this community review report, we discuss applications and techniques for fast machine learning (ML) in science—the concept of integrating powerful ML methods into the real-time experimental data processing loop to accelerate scientific discovery. The material for the report builds on two workshops held by the Fast ML for Science community and covers three main areas: applications for fast ML across a number of scientific domains; techniques for training and implementing performant and resource-efficient ML algorithms; and computing architectures, platforms, and technologies for deploying these algorithms. We also present overlapping challenges across the multiple scientific domains where common solutions can be found. This community report is intended to give plenty of examples and inspiration for scientific discovery through integrated and accelerated ML solutions. This is followed by a high-level overview and organization of technical advances, including an abundance of pointers to source material, which can enable these breakthroughs.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
NASA NTRS1992

Flight simulator for hypersonic vehicle and a study of NASP handling qualities

The research goal of the Human-Machine Systems Engineering Group was to study the existing handling quality studies in aircraft with sonic to supersonic speeds and power in order to understand information requirements needed for a hypersonic vehicle flight simulator. This goal falls within the NASA task statements: (1) develop flight simulator for hypersonic vehicle; (2) study NASP handling qualities; and (3) study effects of flexibility on handling qualities and on control system performance. Following the above statement of work, the group has developed three research strategies. These are: (1) to study existing handling quality studies and the associated aircraft and develop flight simulation data characterization; (2) to develop a profile for flight simulation data acquisition based on objective statement no. 1 above; and (3) to develop a simulator and an embedded expert system platform which can be used in handling quality experiments for hypersonic aircraft/flight simulation training.

Ntuen, Celestine A.↗
NASA NTRSDate not supplied

RadLab: A Comprehensive Database and Graphical and Programming Interfaces for Space Radiation Data

RadLab, a component of the NASA Open Science Data Repository (OSDR), is a database of radiation measurements from multiple instruments and spacecraft that provides visual and programmatic interfaces for interrogation and retrieval of these data. The attributes of data available through RadLab include spacecraft, types of radiation sensing instruments, locations within the spacecraft (e.g. ISS modules), associated celestial bodies, trajectories, and spacecraft coordinates; the primary type of data is the absorbed dose rate, as well as flux and dose equivalent rate where available. The application programming interface (API) implements a request syntax for retrieval of timestamped data filtered by various combinations of such attributes; the graphical user interface (GUI) extends this functionality with visualizations (time series plots, comparison plots, geospatial visualizations) which provide easy means to assess data availability, iteratively refine search parameters, interactively inspect the data, and export target data subsets. Datasets are continuously being added to the RadLab database as part of the rolling release process. Investigators from multiple countries, including the US, Canada, Germany, Bulgaria, Hungary, Italy, Japan, Russia and the Czech Republic, have committed to provide data from their instruments in and beyond low Earth orbit. The current release contains datasets provided by US and international collaborators and includes readings from multiple modules of the ISS, the BioSentinel CubeSat, Chang’e 4, the Lunar Reconnaissance Orbiter, the ExoMars Orbiter, and the Curiosity rover. Datasets are associated with respective RadLab knowledgebase articles which include instrument descriptions and provide bibliographical references. RadLab aims to provide a comprehensive, dynamic compendium of space radiation data, enabling the scientific community to perform analyses of data from multiple detectors and to determine the radiation environment of research missions and experiments. Some of its applications include inference of absorbed radiation dose for NASA GeneLab payloads, and training predictive models as part of the 2024 FDL-X challenge. The platform is actively expanding and seeking additional data, with plans to also cover past (e.g. Shuttle, Mir) and future (e.g. Artemis) missions. The RadLab Working Group has been created to aid in this process as well as to foster collaborations among data contributors and users, to develop standards for data harmonization, and to guide the development of the platform, with the goal to establish the use of RadLab in space radiation research and to advance our understanding of the radiation environment in outer space.

Kirill Grigorev↗
NASA NTRSDate not supplied

RadLab: A Comprehensive Database and Graphical and Programming Interfaces for Space Radiation Data

RadLab, a component of the NASA Open Science Data Repository (OSDR), is a database of radiation measurements from multiple instruments and spacecraft that provides visual and programmatic interfaces for interrogation and retrieval of these data. The attributes of data available through RadLab include spacecraft, types of radiation sensing instruments, locations within the spacecraft (e.g. ISS modules), associated celestial bodies, trajectories, and spacecraft coordinates; the primary type of data is the absorbed dose rate, as well as flux and dose equivalent rate where available. The application programming interface (API) implements a request syntax for retrieval of timestamped data filtered by various combinations of such attributes; the graphical user interface (GUI) extends this functionality with visualizations (time series plots, comparison plots, geospatial visualizations) which provide easy means to assess data availability, iteratively refine search parameters, interactively inspect the data, and export target data subsets. Datasets are continuously being added to the RadLab database as part of the rolling release process. Investigators from multiple countries, including the US, Canada, Germany, Bulgaria, Hungary, Italy, Japan, Russia and the Czech Republic, have committed to provide data from their instruments in and beyond low Earth orbit. The current release contains datasets provided by US and international collaborators and includes readings from multiple modules of the ISS, the BioSentinel CubeSat, Chang’e 4, the Lunar Reconnaissance Orbiter, the ExoMars Orbiter, and the Curiosity rover. Datasets are associated with respective RadLab knowledgebase articles which include instrument descriptions and provide bibliographical references. RadLab aims to provide a comprehensive, dynamic compendium of space radiation data, enabling the scientific community to perform analyses of data from multiple detectors and to determine the radiation environment of research missions and experiments. Some of its applications include inference of absorbed radiation dose for NASA GeneLab payloads, and training predictive models as part of the 2024 FDL-X challenge. The platform is actively expanding and seeking additional data, with plans to also cover past (e.g. Shuttle, Mir) and future (e.g. Artemis) missions. The RadLab Working Group has been created to aid in this process as well as to foster collaborations among data contributors and users, to develop standards for data harmonization, and to guide the development of the platform, with the goal to establish the use of RadLab in space radiation research and to advance our understanding of the radiation environment in outer space.

Kirill Grigorev↗
DOE OSTI2017

ASCR/HEP Exascale Requirements Review Report

This draft report summarizes and details the findings, results, and recommendations derived from the ASCR/HEP Exascale Requirements Review meeting held in June, 2015. The main conclusions are as follows. 1) Larger, more capable computing and data facilities are needed to support HEP science goals in all three frontiers: Energy, Intensity, and Cosmic. The expected scale of the demand at the 2025 timescale is at least two orders of magnitude -- and in some cases greater -- than that available currently. 2) The growth rate of data produced by simulations is overwhelming the current ability, of both facilities and researchers, to store and analyze it. Additional resources and new techniques for data analysis are urgently needed. 3) Data rates and volumes from HEP experimental facilities are also straining the ability to store and analyze large and complex data volumes. Appropriately configured leadership-class facilities can play a transformational role in enabling scientific discovery from these datasets. 4) A close integration of HPC simulation and data analysis will aid greatly in interpreting results from HEP experiments. Such an integration will minimize data movement and facilitate interdependent workflows. 5) Long-range planning between HEP and ASCR will be required to meet HEP's research needs. To best use ASCR HPC resources the experimental HEP program needs a) an established long-term plan for access to ASCR computational and data resources, b) an ability to map workflows onto HPC resources, c) the ability for ASCR facilities to accommodate workflows run by collaborations that can have thousands of individual members, d) to transition codes to the next-generation HPC platforms that will be available at ASCR facilities, e) to build up and train a workforce capable of developing and using simulations and analysis to support HEP scientific research on next-generation systems.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2016

ASCR/HEP Exascale Requirements Review Report

This draft report summarizes and details the findings, results, and recommendations derived from the ASCR/HEP Exascale Requirements Review meeting held in June, 2015. The main conclusions are as follows. 1) Larger, more capable computing and data facilities are needed to support HEP science goals in all three frontiers: Energy, Intensity, and Cosmic. The expected scale of the demand at the 2025 timescale is at least two orders of magnitude -- and in some cases greater -- than that available currently. 2) The growth rate of data produced by simulations is overwhelming the current ability, of both facilities and researchers, to store and analyze it. Additional resources and new techniques for data analysis are urgently needed. 3) Data rates and volumes from HEP experimental facilities are also straining the ability to store and analyze large and complex data volumes. Appropriately configured leadership-class facilities can play a transformational role in enabling scientific discovery from these datasets. 4) A close integration of HPC simulation and data analysis will aid greatly in interpreting results from HEP experiments. Such an integration will minimize data movement and facilitate interdependent workflows. 5) Long-range planning between HEP and ASCR will be required to meet HEP's research needs. To best use ASCR HPC resources the experimental HEP program needs a) an established long-term plan for access to ASCR computational and data resources, b) an ability to map workflows onto HPC resources, c) the ability for ASCR facilities to accommodate workflows run by collaborations that can have thousands of individual members, d) to transition codes to the next-generation HPC platforms that will be available at ASCR facilities, e) to build up and train a workforce capable of developing and using simulations and analysis to support HEP scientific research on next-generation systems.

Habib, Salman↗
DOE OSTI2020

Development of Remote Cutting Tools for use at ATR - 20438

As a result of years of operations of the Advanced Test Reactor (ATR) at the Idaho National Lab (INL), the ATR canal has become congested due to the storage of materials including waste consisting of highly irradiated remote handled (RH) metals. In order to support its continuing nuclear mission the ATR requires canal space to support fuel storage and future core internal change out (CIC) components. INL staff determined that the canal must be cleaned out and waste removed to efficiently utilize the valuable canal floor space. It was determined that a methodology for processing, packaging, characterizing, and removing current and future irradiated hardware from the ATR canal is needed. Orano Federal Services (OFS) and its subcontractor Babcock Services Inc. (BSI) were selected to provide remote tooling in order to safely and efficiently size reduce the material currently stored in the ATR canal. INL developed detailed requirements that established the design and performance requirements for the tooling. Orano TN and BSI had recently successfully completed similar tasks in clearing highly activated materials from spent fuel pools and canals at commercial nuclear plants. That experience directly translated into a proven approach that could be used at the ATR thereby saving time and money and lowering risks. Based on BSI's demonstrated proficiency in the design and deployment of remote size reduction specialty tooling at comparable commercial facilities, Orano Federal Services was confident that a solution to INL's requirements could be delivered. The tooling had to be designed to be compatible with the operational requirements of the ATR canal as well as the functionality needed by the facility operators that will use the equipment. Some of the major design requirements and functions of the equipment included remote operations, capability to shear highly irradiated sections of aluminum and steel tubing, pipe, and components ranging in size from 0.6 cm to 16.5 cm diameter into 1.5 m -1.8 m lengths, utilizing buoyancy compensation to assist with underwater operations, and use commercially available existing technology and components as much as practical. The design team evaluated and selected commercially available equipment that could be modified to meet the design and performance specifications. In addition, a waste sizing table (WST) was designed and fabricated to further ensure the size reduction equipment would perform as expected and give the facility operators a stable and usable remote, underwater work platform. After successful development of the tooling, a mock up test and operator training was conducted to ensure the tooling performed as expected and met the requirements of INL. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2021

Characterizing Impacts of Storage Faults on HPC Applications: A methodology and insights

In recent years, the increasing complexity in scientific simulations and emerging demands for training heavy artificial intelligence models require massive and fast data accesses, which urges high-performance computing (HPC) platforms to equip with more advanced storage infrastructures such as solid-state disks (SSDs). While SSDs offer high-performance I/O, it remains unclear about the reliability challenges faced by the HPC applications under the SSD-related failures, in particular, failures resulting in data corruptions. The goal of this paper is to understand the impact of SSD-related data corruptions on the behaviors of complex HPC applications. To this end, we propose FFIS, a FUSE-based fault injection framework that systematically introduces storage faults into the application layer to model the errors originated from SSDs. FFIS is able to plant different I/O related faults into the data returned from underlying file systems, which also enables the investigation on the error resilience characteristics of the scientific file format for the first time. We demonstrate the use of FFIS with three representative real HPC applications, show how each application reacts to the data corruptions, and provide insights on the error resilience of the widely-adopted HDF5 file format for the HPC applications.

Fang, Bo↗
DOE OSTI2020

Stochasticity and robustness in spiking neural networks

Despite drawing inspiration from biological systems which are inherently noisy and variable, artificial neural networks have been shown to require precise weights to carry out the task which they are trained to accomplish. This creates a challenge when adapting these artificial networks to specialized execution platforms which may encode weights in a manner which restricts their accuracy and/or precision.Reflecting back on the non-idealities which are observed in biological systems, we investigated the effect these properties have on the robustness of spiking neural networks under perturbations to weights. First, we examined techniques extant in conventional neural networks which resemble noisy processes, and postulated they may produce similar beneficial effects in spiking neural networks. Second, we evolved a set of spiking neural networks utilizing biological non-idealities to solve a pole-balancing task, and estimated their robustness. We showed it is higher in networks using noisy neurons, and demonstrated that one of these networks can perform well under the variance expected when a hafnium-oxide based resistive memory is used to encode synaptic weights. Lastly, we trained a series of networks using a surrogate gradient method on the MNIST classification task. We confirmed that these networks demonstrate similar trends in robustness to the evolved networks. We discuss these results and argue that they display empirical evidence supporting the role of noise as a regularizer which can increase network robustness.

97 MATHEMATICS AND COMPUTING↗