SEARCH · Engineering Papers

Results for “shared memory”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 577 records · Page 32

All sources NASA NTRS DOE OSTI

DOE OSTI2026

HARMONY: Large-Scale Architecture Search for Efficient Hybrid Language Models

As large language models scale to trillions of parameters, their computational and memory requirements present critical challenges for efficient training and deployment. While Mixture of Experts (MoE) architectures enable efficient scaling through sparse parameter activation, and state-space models like Mamba offer linear-time complexity, principled methods for combining these paradigms remain undeveloped. We introduce HARMONY (Hybrid Architecture Research for Mamba, Optimized with Neural efficiencY), a multi-objective evolutionary neural architecture search framework for discovering efficient hybrid language models that integrate Transformer attention mechanisms, Mixture-of-Experts routing, and Mamba state-space components. Through large-scale distributed search using 16,384 MI250X GPUs on the Frontier supercomputer, HARMONY explores a comprehensive design space encompassing six attention variants (MHA, MQA, GQA, MLA, SWA, and Mamba-2), variable MoE configurations with both routed and shared experts, and extensive Mamba hyperparameters. Our framework discovers heterogeneous architectures that balance training performance with computational efficiency through multi-objective optimization incorporating latency penalties and fitness-based selection. Analysis of discovered architectures reveals that optimal hybrid designs favor heterogeneous component mixing rather than homogeneous patterns, with Mamba-2 and Multi-Head Latent Attention (MLA) emerging as preferred mechanisms. Discovered architectures demonstrate superior training efficiency: our best configuration achieves a final perplexity of 1.0874 with 2.38B parameters while processing 4,320 tokens/second, outperforming significantly larger manually designed models. Full-scale evaluation shows HARMONY's top architectures achieve better loss trajectories than equivalently-sized models using state-of-the-art configurations including Mixtral, Jamba, and Samba. Additionally, we demonstrate 91% weak scaling efficiency when training discovered 36B-parameter models across 1,024 GPUs. HARMONY is released as an open framework with comprehensive tools for building and training hybrid models using expert-data-pipeline parallelism, democratizing access to automated architecture design for next-generation language models.

Herron, Emily [ORNL] (ORCID:0000000273008172)↗

NASA NTRS2005

ASK Magazine; No. 21

THIS ISSUE FEATURES A VISUAL DEPICTION OF THE ACADEMY of Program and Project Leadership (APPL). I imagine a variety of initial reactions to the drawing. One might be, "What is a cartoon doing in a magazine about project management?" Or perhaps, "Wow, nice colors-and fun." Another may be to closely search the image for signs, symbols and meaning. Still another, to read a new level of innovation and creativity into the picture. Undoubtedly, some readers will raise questions about the cost. Of course, any reaction is a sign of engagement. The stronger, the more energized the emotional and cognitive processing, the better. It is a sign of attention and interaction. For I've heard it said, "You only need to worry if they don t care one way or the other." So what is the point of the picture? To stimulate interest, raise questions, promote discussion, and maybe raise a smile.. .That, at least, was my initial reaction when I was introduced to the work of Nancy Hegedus, who helps to create these drawings for Root Learning Inc. At the NASA PM Conference, I was first shown the work Nancy had been doing with the help of Goddard s Knowledge Management Architect, Dr. Ed Rogers. I was immediately drawn into the power of visualization as a tool for more effective learning, communicating, and conveying complex knowledge concepts. We need new tools in today s world, where information and data overwhelms by sheer volume. There are articles, pamphlets, communications, and white papers-all aiming to convince and influence. Reactions to these tend to be either avoidance or mind-numbing, heavy-eyed consent; the message never registers or enters the soul. That s one of the reasons that APPL s Knowledge Sharing Initiative (KSI) has turned to storytelling as a memorable way of transfer- ring knowledge, inspiring imitation of best practices, and spurring reflection. ASK Magazine s recent fourth birthday marks an important milestone in APPL s continuing quest to provide ongoing support to project managers and to promote mission success. And similar to storytelling, the power of visualization is receiving increasing attention in recent years as a way to stimulate engagement. Pictures and visual graphs are viewed as one of the most effective ways for displaying, describing, and generating discussion about quantitative and technically complex information. Prototypes, models, and simulations are considered essential for stimulating innovation through open and engaging discussions. There has also been extensive writing on the use of visual graphics, pictures, and cartoons to facilitate memory, creativity, openness, attention-and even well-being. For many of these reasons, I am excited to have a colorful visual depiction of the APPL world included in ASK. Without the addition of text or slides, the intent is to invite people into the world of the APPL mission-as well as its products, services, customers, and partners- in a fun and engaging manner. As project leaders strive to find ways to encourage engagement, learning, and transmission of knowledge, traditional technologies are proving to be as valuable as modern technologies. (But for those who want more information in the form of texts and slide presentations, we certainly have an abundance of those as well.)

Laufer, Alexander↗

NASA NTRS1988

Networking and AI systems: Requirements and benefits

The price performance benefits of network systems is well documented. The ability to share expensive resources sold timesharing for mainframes, department clusters of minicomputers, and now local area networks of workstations and servers. In the process, other fundamental system requirements emerged. These have now been generalized with open system requirements for hardware, software, applications and tools. The ability to interconnect a variety of vendor products has led to a specification of interfaces that allow new techniques to extend existing systems for new and exciting applications. As an example of the message passing system, local area networks provide a testbed for many of the issues addressed by future concurrent architectures: synchronization, load balancing, fault tolerance and scalability. Gold Hill has been working with a number of vendors on distributed architectures that range from a network of workstations to a hypercube of microprocessors with distributed memory. Results from early applications are promising both for performance and scalability.

Source record↗

NASA NTRS1994

Flexible Method for Inter-object Communication in C++

A method has been developed for organizing and sharing large amounts of information between objects in C++ code. This method uses a set of object classes to define variables and group them into tables. The variable tables presented here provide a convenient way of defining and cataloging data, as well as a user-friendly input/output system, a standardized set of access functions, mechanisms for ensuring data integrity, methods for interprocessor data transfer, and an interpretive language for programming relationships between parameters. The object-oriented nature of these variable tables enables the use of multiple data types, each with unique attributes and behavior. Because each variable provides its own access methods, redundant table lookup functions can be bypassed, thus decreasing access times while maintaining data integrity. In addition, a method for automatic reference counting was developed to manage memory safely.

Curlett, Brian P.↗

NASA NTRS2009

Economical Implementation of a Filter Engine in an FPGA

A logic design has been conceived for a field-programmable gate array (FPGA) that would implement a complex system of multiple digital state-space filters. The main innovative aspect of this design lies in providing for reuse of parts of the FPGA hardware to perform different parts of the filter computations at different times, in such a manner as to enable the timely performance of all required computations in the face of limitations on available FPGA hardware resources. The implementation of the digital state-space filter involves matrix vector multiplications, which, in the absence of the present innovation, would ordinarily necessitate some multiplexing of vector elements and/or routing of data flows along multiple paths. The design concept calls for implementing vector registers as shift registers to simplify operand access to multipliers and accumulators, obviating both multiplexing and routing of data along multiple paths. Each vector register would be reused for different parts of a calculation. Outputs would always be drawn from the same register, and inputs would always be loaded into the same register. A simple state machine would control each filter. The output of a given filter would be passed to the next filter, accompanied by a "valid" signal, which would start the state machine of the next filter. Multiple filter modules would share a multiplication/accumulation arithmetic unit. The filter computations would be timed by use of a clock having a frequency high enough, relative to the input and output data rate, to provide enough cycles for matrix and vector arithmetic operations. This design concept could prove beneficial in numerous applications in which digital filters are used and/or vectors are multiplied by coefficient matrices. Examples of such applications include general signal processing, filtering of signals in control systems, processing of geophysical measurements, and medical imaging. For these and other applications, it could be advantageous to combine compact FPGA digital filter implementations with other application-specific logic implementations on single integrated-circuit chips. An FPGA could readily be tailored to implement a variety of filters because the filter coefficients would be loaded into memory at startup.

Kowalski, James E.↗

DOE OSTI2021

Bulk Glassy GeTe 2 : A Missing Member of the Tetrahedral GeX 2 Family and a Precursor for the Next Generation of Phase-Change Materials

Vitreous germanium disulfide GeS 2 and diselenide GeSe 2 belong to canonical chalcogenide glasses extensively studied over the past half century. Their high-temperature orthorhombic polymorphs are congruently melting compounds, and the tetrahedral crystal and glass structure is largely preserved in the melt. In contrast, the ditelluride counterpart is absent in the Ge-Te phase diagram, which shows only a single compound, monotelluride GeTe. Phase-change materials based on GeTe have become a technologically important class of solids, and their structure and properties are also widely studied. Surprisingly, very scarce information is available for alloys having GeTe 2 stoichiometry. Using a fast quenching procedure in silica capillaries, high-energy X-ray diffraction, and Raman spectroscopy supported by first-principles simulations, we show that bulk glassy GeTe 2 differs substantially from the lighter GeX 2 members, revealing 46% of trigonal germanium, 31% of three-fold coordinated tellurium, and only 20% of edge-sharing tetrahedra or pyramids. The fraction of homopolar Ge-Ge bonds is low; however, the population of dominant Te-Te dimers and Te n oligomers, n <= 10, appears to be significant. The complex structural and chemical topology of g-GeTe 2 is directly related to the thermodynamic metastability of germanium ditelluride, schematically represented by the following reaction: GeTe 2 $\rightleftarrows$ GeTe + Te. Disproportionation is complete above liquidus in the temperature range of semiconductor-metal transition, and the dense metallic GeTe 2 liquid, mostly consisting of five-fold coordinated Ge species, exhibits high fluidity, strong fragility (m = 99 ± 5), and presumably a fast structural transformation rate combined with low atomic mobility in the vicinity of the glass transition temperature, favorable for reliable long-term data retention in nonvolatile memories. The observed and predicted characteristic features make GeTe 2 a promising precursor for the next generation of phase-change materials, especially coupled with additional metal doping, depolymerizing the tetrahedral interconnected glass network and accelerating (sub)nanosecond crystallization.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2020

OpenACC unified programming environment for GPU and FPGA multi-hybrid acceleration

Attached accelerators have been frequently used in recent High Per- formance Computing (HPC) systems because of their high performance/power ratio. In particular, the Graphics Processing Unit (GPU) is the most popu- lar accelerator owing to its high peak FLOPS performance and high memory bandwidth supported by HBM2, etc. However, the performance of GPU depends highly on a large degree of SIMD parallelism and has difficulty sustaining a high performance on programs with frequent branch operations or a partially low degree of parallelism.By contrast, a Field Programmable Gate Array (FPGA) has received attention as a different type of accelerator than GPU as a fully reconfigurable processor fitting the target applications. The high performance of FPGA is mainly provided by a pipelined operation and optimized circuit suitable for any operation even with frequent conditional branches. We have been focusing on the flexibility of FPGA to compensate for the weakness of GPU. We believe that the coupling of GPU with FPGA can result in one of the most powerful accelerating platforms available.However, the program coding of GPU and FPGA coupling can be quite difficult for application users. Traditionally, CUDA by NVIDIA has been the most popular programming language with the largest share of GPUs used in HPC, whereas a hardware description language such as Verilog HDL has been used in FPGA programming. OpenCL coding has recently become available even on high-end FPGAs. Moreover, several recent studies have also enabled the OpenACC coding for use in FPGA. In this study, we provide a unified programming system based on OpenACC for a platform equipped with both GPU and FPGA aiming at the next-generation accelerated supercomputer framework. Our programming environment is called Multi-Hybrid OpenACC Translator (MHOAT), and in this paper, we describe the basic concept and prototype system of MHOAT based on an evaluation on the amount of coding required and the performance of a hybrid multi-device accelerated system.

Tsunashima, Ryuta↗

DOE OSTI2025

Electrochemical Random-Access Memory: Progress, Perspectives, and Opportunities

Non-von Neumann computing using neuromorphic systems based on analogue synaptic and neuronal elements has emerged as a potential solution to tackle the growing need for more efficient data processing, but progress toward practical systems has been stymied due to a lack of materials and devices with the appropriate attributes. Recently, solid state electrochemical ion-insertion, also known as electrochemical random access memory (ECRAM) has emerged as a promising approach to realize the needed device characteristics. ECRAM is a three terminal device that operates by tuning electronic conductance in functional materials through solid-state electrochemical redox reactions. This mechanism can be considered as a gate-controlled bulk modulation of dopants and/or phases in the channel. Early work demonstrating that ECRAM can achieve nearly ideal analogue synaptic characteristics has sparked tremendous interest in this approach. More recently, the realization that electrochemical ion insertion can be used to tune the electronic properties of many types of materials including transition metal oxides, layered two-dimensional materials, organic and coordination polymers, and that the changes in conductance can span orders of magnitude has further attracted interest in ECRAM as the basis for analogue synaptic elements for inference accelerators as well as for dynamical devices that can emulate a wide range of neuronal characteristics for implementation in analogue spiking neural networks. At its core, ECRAM shares many fundamental aspects with rechargeable batteries, where ion insertion materials are used extensively for their ability to reversibly store charge and energy. Computing applications, however, present drastically different requirements: systems will require many millions of devices, scaled down to tens of nanometers, all while achieving reliable electronic-state tuning at scaled-up rates and endurances, and with minimal energy dissipation and noise. Further, in this review, we discuss the history, basic concepts, recent progress, as well as the challenges and opportunities for different types of ECRAM, broadly grouped by their primary mobile ionic charge carrier, including Li, protons, and oxygen vacancies.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

NASA NTRS2000

Interference Lattice-based Loop Nest Tilings for Stencil Computations

A common method for improving performance of stencil operations on structured multi-dimensional discretization grids is loop tiling. Tile shapes and sizes are usually determined heuristically, based on the size of the primary data cache. We provide a lower bound on the numbers of cache misses that must be incurred by any tiling, and a close achievable bound using a particular tiling based on the grid interference lattice. The latter tiling is used to derive highly efficient loop orderings. The total number of cache misses of a code is the sum of (necessary) cold misses and misses caused by elements being dropped from the cache between successive loads (replacement misses). Maximizing temporal locality is equivalent to minimizing replacement misses. Temporal locality of loop nests implementing stencil operations is optimized by tilings that avoid data conflicts. We divide the loop nest iteration space into conflict-free tiles, derived from the cache miss equation. The tiling involves the definition of the grid interference lattice an equivalence class of grid points whose images in main memory map to the same location in the cache-and the construction of a special basis for the lattice. Conflicts only occur on the boundaries of the tiles, unless the tiles are too thin. We show that the surface area of the tiles is bounded for grids of any dimensionality, and for caches of any associativity, provided the eccentricity of the fundamental parallelepiped (the tile spanned by the basis) of the lattice is bounded. Eccentricity is determined by two factors, aspect ratio and skewness. The aspect ratio of the parallelepiped can be bounded by appropriate array padding. The skewness can be bounded by the choice of a proper basis. Combining these two strategies ensures that pathologically thin tiles are avoided. They do not, however, minimize replacement misses per se. The reason is that tile visitation order influences the number of data conflicts on the tile boundaries. If two adjacent tiles are visited successively, there will be no replacement misses on the shared boundary. The iteration space may be covered with pencils larger than the size of the cache while avoiding data conflicts if the pencils are traversed by a scanning-face method. Replacement misses are incurred only on the boundaries of the pencils, and the number of misses is minimized by maximizing the volume of the scanning face, not the volume of the tile. We present an algorithm for constructing the most efficient scanning face for a given grid and stencil operator. In two dimensions it is based on a continued fraction algorithm. In three dimensions it follows Voronoi's successive minima algorithm. We show experimental results of using the scanning face, and compare with canonical loop orderings.

VanderWijngaart, Rob F.↗

DOE OSTI2020

Ten-Year Review of US/German Collaboration in Salt Repository Research - 20069

A multinational collaboration on salt repository research, design and operation has enjoyed remarkable success since reinvigorating efforts in 2010. Nations now engaged in the shared salt repository research agenda include Germany, the United States, the United Kingdom, the Netherlands, and Poland. The scientific basis for safe and permanent disposal of nuclear waste in salt formations has been strengthened by annual workshops that recognize and address contemporary research, including breakout sessions to stimulate open discussion and focus planning for ongoing investigations. Collaboration not only identifies pertinent technical issues, but facilitates timely, expert, and cost-effective consideration. Contemporary workshops have been held annually since 2010 and are documented in yearly state-of-the-art Proceedings, which summarize content and conclusions. The Proceedings help preserve scientific understanding and provide timely source references. These workshops often produce valuable joint publications coordinated with the Nuclear Energy Agency and other suitable external forums for dissemination. Nuclear waste management programs face growing challenges, while permanent disposal in salt formations provides a robust, safe option for several nations. Workshop format and publications provide a cost-effective insurance against loss of scientific expertise and institutional memory. This paper summarizes ten US/German workshops since formal reinitiation, reexamines key technical issues, discusses the evolving research agenda, and highlights successes and challenges. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

DOE OSTI2021

$\mathrm{PPT}$-Multicore: performance prediction of Open$\mathrm{MP}$ applications using reuse profiles and analytical modeling

In this report we present PPT-Multicore, an analytical model embedded in the Performance Prediction Toolkit (PPT) to predict parallel applications’ performance running on a multicore processor. PPT-Multicore builds upon our previous work towards a multicore cache model. We extract LLVM basic block labeled memory trace using an architecture-independent LLVM-based instrumentation tool only once in an application’s lifetime. The model uses the memory trace and other parameters from an instrumented sequentially executed binary. We use probabilistic and computationally efficient reuse profiles to predict the cache hit rates and runtimes of OpenMP programs’ parallel sections. We model Intel’s Broadwell, Haswell, and AMD’s Zen2 architectures and validate our framework using different applications from PolyBench and PARSEC benchmark suites. The results show that PPT-Multicore can predict cache hit rates with an overall average error rate of 1.23% while predicting the runtime with an error rate of 9.08%.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2025

Braiding for the win: Harnessing braiding statistics in topological states to play quantum games

Nonlocal quantum games provide proof of principle that quantum resources can confer an advantage at certain tasks. They also provide a compelling way to explore the computational utility of phases of matter on quantum hardware. In a recent paper [O. Hart et al., Phys. Rev. Lett. 134, 130602 (2025)], we demonstrated that a toric code resource state conferred advantage at a certain nonlocal game, which remained robust to small deformations of the resource state. In this paper we demonstrate that this robust advantage is a generic property of resource states drawn from topological or fracton ordered phases of quantum matter. To this end, we illustrate how several other states from paradigmatic topological and fracton ordered phases can function as resources for suitably defined nonlocal games, notably the three-dimensional toric-code phase, the X-cube fracton phase, and the double-semion phase. The key in every case is to design a nonlocal game that harnesses the characteristic braiding processes of a quantum phase as a source of contextuality. We unify the strategies that take advantage of mutual statistics by relating the operators to be measured to order and disorder parameters of an underlying generalized symmetry-breaking phase transition. Additionally, by connecting the win probability to twist products, we show that success at the game serves as a many-body entanglement witness. Namely, if the players implement a perfect quantum strategy on large length scales, the quantum state they share cannot be connected to a trivial product state via a constant-depth local unitary circuit. Lastly, we massively generalize the family of games that admit perfect strategies when codewords of homological quantum error-correcting codes are used as resources.

Fractons↗

NASA NTRSDate not supplied

Using XR for Improving Scientific Discovery With Numerical Weather Models

Earth science (ES) digital twins will help us understand the complex interactions and interrelationships that make up our Earth system and the impacts of earth science phenomena on it. Our work addresses two underdeveloped areas in current ES digital twin work: improving the understanding and interaction with ES model outputs by using Virtual and Mixed Reality (XR) tools and improving the non-intuitive mapping of continuous ES natural phenomena to gridded reference frames in current numerical models. Traditionally, scientists working on ES view and analyze the results of calculated or measured observables with static 1-dimensional (1D), 2D or 3D plots displayed on flat computer screens or paper. Using such limited mediums, it can be very difficult to identify, track and understand the evolution of key features due to poor viewing angles and the nature of flat computer screens. In addition, numerical models, such as the NASA Goddard Earth Observing System (GEOS) ES model, are almost exclusively formulated, visualized and analyzed in an Eulerian reference frame with fixed grid points in space and time. However, ES phenomena such as convective clouds, hurricanes and wildfire smoke plumes are visualized and analyzed in a Lagrangian reference frame: therefore it is often difficult and unnatural to understand these phenomena in relation to each other, visualized either in an Eulerian or Lagrangian context. In 3D visualizations, data generally takes one of three forms: gridded (e.g., voxelized) data, where space is divided into regions; point clouds, where data is represented as a set of points; and meshes, where objects are rendered as surfaces composed of small polygons (usually triangles). A gridded, Eulerian reference frame has been the default representation for the 2D visual analysis of atmospheric data in part because the numerical methods used to generate atmospheric model data in the first place use a gridded approach, with equations defining the relationships between the physical variables in each of a grid's cells across successive timesteps. In our work, we are particularly interested in data from GEOS. Another reason why gridded representations tend to be used for visualizing data from such models is because trajectories are difficult to interpret from representations on 2D surfaces, due to line-of-sight ambiguity. Instead of a fixed grid from GEOS, we embed a trajectory model to simulate particles' movement throughout a GEOS run. We then ingest these particle trajectories as animated point clouds with a NASA open source XR toolkit, the Mixed Reality Exploration Toolkit (MRET), and merge GEOS data with ES phenomena data onto one combined visualization that the user can intuitively interact with. Efficient rendering of arbitrarily large point clouds is an ongoing challenge being addressed by the computer science community, with the GPU-based optimizations and efficient GPU memory utilization a common theme of recent advances, especially for XR, where sustained high frame rate is mandatory to save the user from suffering due to simulation sickness. In this work, we describe and evaluate our progress in choosing and implementing appropriate methods for rendering arbitrarily large point clouds within MRET for XR. While tracking the XR headset enables the immersion of a user within a 3D scene of a data visualization, tracking of XR handheld controllers or user’s hands enables us to implement intuitive user interactions with the visualized datasets. Conventional tools require a user working with an ES visualization to conduct many interactions to commit their intended selections or manipulations with a visualized dataset; for example to specify a set of points in 3D space. Doing so in a 2D flat screen interface has traditionally required specifying a set of points in three distinct 2D coordinate systems (XY, XZ, and YZ), which is cumbersome. In other scientific domains, it has been shown that specifying or selecting a location or volume in XR using handheld controllers or tracked hands allows for greater speed and accuracy. We anticipate the same will hold true for atmospheric data, and we will share initial results of measuring the utility of such an interface. Notably, as the data being visualized is generated by GEOS as a prediction based on initial conditions, an intended application of our tool is to serve as part of an iterative feedback loop. Through XR, a scientist will review and manipulate a GEOS model run, modifying the conditions as needed to do subsequent runs of GEOS. Thereby, XR-based improvements to speed and accuracy of 3D tagging of points minimizes the effort required by both the scientist and the computer cluster conducting the necessary calculations.

Thomas Grubb↗

DOE OSTI2021

Cryogenic Vibrationally Resolved Photoelectron Spectroscopy of OH–(H2O): Confirmation of Multidimensional Franck-Condon Simulation Results for the Transition State of the OH + H2O Reaction

We present a transition state spectroscopic study of the OH + H2O reaction using the experimental technique of cryogenic Negative Ion Photoelectron Spectroscopy (NIPES). The recorded NIPE spectrum at 193 nm exhibits multiple vibrational progressions that include excitations to the shared H atom antisymmetric stretching mode with an interval of 0.32 eV as well as other progressions, mainly involving the H bending and O…O symmetric stretching modes. The Vertical Detachment Energy (VDE) was measured at 3.53 eV, whereas an upper limit for the Adiabatic Detachment Energy (ADE) was estimated at 2.90 eV. These values are in excellent agreement with the theoretically computed values of 3.51 eV and 2.83 eV respectively, obtained at the CCSD(T)/aug-cc-pV5Z level of theory. The recorded NIPE spectrum is in very good agreement when compared to the one recently reported from four-dimensional Franck-Condon simulations, in which a similar spectral profile was predicted. Besides observing the ground state, a charge-transfer excited state in the form of [OH–(H2O)+] is identified with a relative energy of 1.39 eV, well matching the previous prediction of 1.36 eV. This work was supported by U.S. Department of Energy (DOE), Office of Science, Office of Basic Energy Sciences, Division of Chemical Sciences, Geosciences, and Biosciences, and performed using EMSL, a national scientific user facility sponsored by DOE’s Office of Biological and Environmental Research and located at Pacific Northwest National Laboratory, which is operated by Battelle Memorial Institute for the DOE. The theoretical calculations were conducted on EMSL’s “Cascade” Supercomputer. This research also used resources of the National Energy Research Scientific Computing Center, which is supported by the Office of Science of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231.

Cao, Wenjin↗

DOE OSTI2021

Deep Cellular Recurrent Network for Efficient Analysis of Time-Series Data With Spatial Information

Efficient processing of large-scale time series data is an intricate problem in machine learning. Conventional sensor signal processing pipelines with hand engineered feature extraction often involve huge computational cost with high dimensional data. Deep recurrent neural networks have shown promise in automated feature learning for improved time-series processing. However, generic deep recurrent models grow in scale and depth with increased complexity of the data. This is particularly challenging in presence of high dimensional data with temporal and spatial characteristics. Consequently, this work proposes a novel deep cellular recurrent neural network (DCRNN) architecture to efficiently process complex multi-dimensional time series data with spatial information. Here, the cellular recurrent architecture in the proposed model allows for location-aware synchronous processing of time series data from spatially distributed sensor signal sources. Extensive trainable parameter sharing due to cellularity in the proposed architecture ensures efficiency in the use of recurrent processing units with high-dimensional inputs. This study also investigates the versatility of the proposed DCRNN model for classification of multi-class time series data from different application domains. Consequently, the proposed DCRNN architecture is evaluated using two time-series datasets: a multichannel scalp EEG dataset for seizure detection, and a machine fault detection dataset obtained in-house. The results suggest that the proposed architecture achieves state-of-the-art performance while utilizing substantially less trainable parameters when compared to comparable methods in the literature.

60 APPLIED LIFE SCIENCES↗

NASA NTRS2013

Miga Aero Actuator and 2D Machined Mechanical Binary Latch

Shape memory alloy (SMA) actuators provide the highest force-to-weight ratio of any known actuator. They can be designed for a wide variety of form factors from flat, thin packages, to form-matching packages for existing actuators. SMA actuators can be operated many thousands of times, so that ground testing is possible. Actuation speed can be accurately controlled from milliseconds to position and hold, and even electronic velocity-profile control is possible. SMA actuators provide a high degree of operational flexibility, and are truly smart actuators capable of being accurately controlled by onboard microprocessors across a wide range of voltages. The Miga Aero actuator is a SMA actuator designed specifically for spaceflight applications. Providing 13 mm of stroke with either 20- or 40-N output force in two different models, the Aero actuator is made from low-outgassing PEEK (polyether ether ketone) plastic, stainless steel, and nickel-titanium SMA wires. The modular actuator weighs less than 28 grams. The dorsal output attachment allows the Aero to be used in either PUSH or PULL modes by inverting the mounting orientation. The SPA1 actuator utilizes commercially available SMA actuator wire to provide 3/8-in. (approx. =.1 cm) of stroke at a force of over 28 lb (approx. = .125 N). The force is provided by a unique packaging of the single SMA wire that provides the output force of four SMA wires mechanically in parallel. The output load is shared by allowing the SMA wire to slip around the output attachment end to adjust or balance the load, preventing any individual wire segment from experiencing high loads during actuation. A built-in end limit switch prevents overheating of the SMA element following actuation when used in conjunction with the Miga Analog Driver [a simple MOSFET (metal oxide semiconductor field-effect transistor) switching circuit]. A simple 2D machined mechanical binary latch has been developed to complement the capabilities of SMA wire actuators. SMA actuators typically perform ideally as latch-release devices, wherein a spring-loaded device is released when the SMA actuator actuates in one direction. But many applications require cycling between two latched states open and closed.

Gummin, Mark A.↗

DOE OSTI2022

Cabana: A Performance Portable Library for Particle-Based Simulations

Particle-based simulations are ubiquitous throughout many fields of computational science and engineering, spanning the atomistic level with molecular dynamics (MD), to mesoscale particle-in-cell (PIC) simulations for solid mechanics, device-scale modeling with PIC methods for plasma physics, and massive N-body cosmology simulations of galaxy structures, with many other methods in between (Hockney & Eastwood, 1989). While these methods use particles to represent significantly different entities with completely different physical models, many low-level details are shared including performant algorithms for short- and/or long-range particle interactions, multi-node particle communication patterns, and other data management tasks such as particle sorting and neighbor list construction. Cabana is a performance portable library for particle-based simulations, developed as part of the Co-Design Center for Particle Applications (CoPA) within the Exascale Computing Project (ECP) (Alexander et al., 2020). The CoPA project and its full development scope, including ECP partner applications, algorithm development, and similar software libraries for quantum MD, is described in (Mniszewski et al., 2021). Cabana uses the Kokkos library for on-node parallelism (Edwards et al., 2014; Trott et al., 2022), enabling simulation on multi-core CPU and GPU architectures, and MPI for GPU-aware, multi-node communication. Cabana provides particle simulation capabilities on almost all current Kokkos backends, including serial execution, OpenMP (including OpenMP-Target for GPUs), CUDA (NVIDIA GPUs), HIP (AMD GPUs), and SYCL (Intel GPUs), providing a clear path for the coming generation of accelerator-based exascale hardware. Cabana builds on Kokkos by providing new particle data structures and particle algorithms resulting in a similar execution policy-based, node-level programming model that is intended to be used in addition to the core Kokkos library within an application. Cabana is designed as an application and physics agnostic, but particle-specific toolkit which can either be used to generate a new application, or to be used as needed in existing applications at various levels of invasiveness including through interfaces that wrap user memory in existing data structures.

97 MATHEMATICS AND COMPUTING↗

NASA NTRS2019

Electra: A Modular-Based Expansion of NASA's Supercomputing Capability

NASA has increasingly relied on high-performance computing (HPC) re- sources for computational modeling, simulation, and data analysis to meet the science and engineering goals of its missions in space exploration, aeronautics, and Earth and space science. The NASA Advanced Supercomputing (NAS) Division at Ames Research Center in Silicon Valley, Calif., hosts NASA’s premier supercomputing resources, integral to achieving and enhancing the success of the agency’s missions. NAS provides a balanced environment, funded under the High-End Computing Capability (HECC) project, comprised of world-class supercomputers, including its flagship distributed-memory cluster, Pleiades; high-speed networking; and massive data storage facilities, along with multi-disciplinary support teams for user support, code porting and optimization, and large-scale data analysis and scientific visualization. However, as scientists have increased the fidelity of their simulations and engineers are conducting larger parameter-space studies, the requirements for supercomputing resources have been growing by leaps and bounds. With the facility housing the HECC systems reaching its power and cooling capacity, NAS undertook a prototype project to investigate an alternative approach for housing supercomputers. Modular supercomputing, or container-based computing, is an innovative concept for expanding NASA’s HPC capabilities. With modular supercomputing, additional containers—similar to portable storage pods—can be connected together as needed to accommodate the agency’s ever-increasing demand for computing resources. In addition, taking advantage of the local weather permits the use of cooling technologies that would additionally save energy and reduce annual water usage. The first stage of NASA’s Modular Supercomputing Facility (MSF) prototype, which resulted in a 1,000 square-foot module on a concrete pad with room for 16 compute racks, was completed in Fall 2016 and an SGI (now HPE) computer system, named Electra, was deployed there in early 2017. Cooling is performed via an evaporative system built into the module, and preliminary experience shows a Power Usage Effectiveness (PUE) measurement of 1.03. Electra achieved over a petaflop on the LINPACK benchmark, sufficient to rank number 96 on the November 2016 TOP500 list [14]. The system consists of 1,152 InfiniBand-connected Intel Xeon Broadwell-based nodes. Its users access their files on a facility-wide file system shared by all HECC compute assets via Mellanox MetroX InfiniBand extenders, which connect the Electra fabric to Lustre routers in the primary facility over fiber-optic links about 900 feet long. The MSF prototype has exceeded expectations and is serving as a blueprint for future expansions. In the remainder of this chapter, we detail how modular data center technology can be used to expand an existing compute resource. We begin by describing NASA’s requirements for supercomputing and how resources were provided prior to the integration of the Electra module-based system.

Biswas, Rupak↗