SEARCH · Engineering Papers
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.
Among the many techniques in computer graphics, ray tracing is prized because it can render realistic images, albeit at great computational expense. In this note, the performance of several approaches to ray tracing on a distributed memory parallel system is evaluated. A set of performance instrumentation tools and their associated visualization software are used to identify the underlying causes of performance differences.
Resistive switching in correlated complex oxides is lucrative for emerging applications in neuromorphic computing, and densely scaled non-volatile memory. Electrical conductance of such complex oxides can be controllable switched across multiple orders of magnitude by either (a) electroforming a conduction channel (e.g., in tungsten oxide), or (b) inducing Mott-Hubbard transition (e.g., in rare-earth nickelates)– both via controlled migration of defects (such as oxygen vacancies) under applied bias. Nevertheless, the promise of such defect-driven electronic transitions are far from realized due to a lack of fundamental understanding of the atomic-scale processes that underlie migration and spatiotemporal evolution of oxygen vacancies over nano-to-mesoscopic length/timescales under applied electric field. In this project, we employ a synergistic integration of density functional theory (DFT) calculations, ab initio/classical molecular dynamics (AIMD/CMD) simulations, machine learning (ML), precision synthesis, and multi-modal X-ray imaging experiments to address this knowledge gap. Such an integrated approach offers to elucidate the correlations between subtle structural distortion and oxidation states; treat localized charge carriers; describe defect/ion transport in the presence of electric field; and, in turn, greatly advance the current understanding of microstructural evolution in complex oxides under applied bias. The fundamental knowledge gained from this work will enable precise control over hierarchical defect structures and unravel new routes to manipulate resistance states in complex oxides. This, in turn, will accelerate design of novel devices with desired set of neural functionalities, and high-speed densely-scaled resistive random access memory technologies.
Special test equipment and techniques to collect and process image information from charge coupled devices (CCDs) by digital computer were reviewed. The video channel was traced from the CCD to the direct memory access bus of the Interdata Computer. Software was developed to evaluate and characterize a CCD for (1) dark signal versus temperature relationship, (2) calculation of temporal noise magnitude and noise shape for each pixel, (3) spatial noise into the video chain due to dark signal, (4) response versus illumination relationship (gamma), (5) response versus wavelength of illumination (spectral), (6) optimization of forcing functions, and (7) evaluation of an image viewed by a CCD. The basic software differences and specific examples of each program operating on real data are presented.
The ramp-reversal memory (RRM) effect in metal–insulator transition metal oxides (TMOs), a non-volatile resistance change induced by repeated temperature cycling, has attracted considerable interest in neuromorphic computing and non-volatile memory devices. Our previous defect motion model successfully explained RRM in vanadium dioxide (VO 2 ), capturing observed critical temperature shifts and memory accumulation throughout the sample. However, this approach lacked interactions between metallic and insulating domains. Here, we extend our model by combining a correlated Random Field Ising Model with defect diffusion-segregation, enabling accurate hysteresis modeling while predicting the relationship between RRM and domain interactions. Our simulations demonstrate that the maximum RRM occurs when the turnaround temperature approaches the inflection point. This peak in RRM vs. turnaround temperature is consistent with prior transport measurements, as well as our own optical measurements reported here. Significantly, we find that increasing nearest-neighbor interactions enhances the maximum memory effect, thus providing a clear mechanism for optimizing RRM performance. Since our model employs minimal assumptions, we predict that RRM should be a widespread phenomenon in materials exhibiting patterned phase coexistence of electronic domains. This work not only advances fundamental understanding of memory behavior in TMOs but also establishes a much-needed theoretical framework for optimizing device applications.
Storage backends of parallel compute clusters are still based mostly on magnetic disks, while newer and faster storage technologies such as flash-based SSDs or non-volatile random access memory (NVRAM) are deployed within compute nodes. Including these new storage technologies into scientific workflows is unfortunately today a mostly manual task, and most scientists therefore do not take advantage of the faster storage media. One approach to systematically include nodelocal SSDs or NVRAMs into scientific workflows is to deploy ad hoc file systems over a set of compute nodes, which serve as temporary storage systems for single applications or longer-running campaigns. This paper presents results from the Dagstuhl Seminar 17202 “Challenges and Opportunities of User-Level File Systems for HPC” and discusses application scenarios as well as design strategies for ad hoc file systems using node-local storage media. The discussion includes open research questions, such as how to couple ad hoc file systems with the batch scheduling environment and how to schedule stage-in and stage-out processes of data between the storage backend and the ad hoc file systems. Also presented are strategies to build ad hoc file systems by using reusable components for networking and how to improve storage device compatibility. Various interfaces and semantics are presented, for example those used by the three ad hoc file systems BeeOND, GekkoFS, and BurstFS. Their presentation covers a range from file systems running in production to cutting-edge research focusing on reaching the performance limits of the underlying devices.
A large number of Intelligent Tutoring Systems (ITS's) have been built since they were first proposed in the early 1970's. Research conducted on the use of the best of these systems has demonstrated their effectiveness in tutoring in selected domains. Computer Sciences Corporation, Applied Technology Division, Houston Operations has been tasked by the Spacecraft Software Division at NASA/Johnson Space Center (NASA/JSC) to develop a number of lTS's in a variety of domains and on many different platforms. This paper will address issues facing the development of an ITS on a personal computer using the CLIPS (C Language Integrated Production System) language. For an ITS to be widely accepted, not only must it be effective, flexible, and very responsive, it must also be capable of functioning on readily available computers. There are many issues to consider when using CLIPS to develop an ITS on a personal computer. Some of these issues are the following: when to use CLIPS and when to use a procedural language such as C, how to maximize speed and minimize memory usage, and how to decrease the time required to load your rule base once you are ready to deliver the system. Based on experiences in developing the CLIPS Intelligent Tutoring System (CLIPSITS) on an IBM PC clone and an intelligent Physics Tutor on a Macintosh 2, this paper reports results on how to address some of these issues. It also suggests approaches for maintaining a powerful learning environment while delivering robust performance within the speed and memory constraints of the personal computer.
Graph algorithms with polynomial space and time requirements often become infeasible for massive graphs with billions of edges or more. State-of-the-art approaches therefore employ approximate serial, parallel, and distributed algorithms to tackle these challenges. However, such approaches require storing the entire graph in memory and thus need access to costly computing resources such as clusters and supercomputers. In this paper, we present practical streaming approaches for solving massive graph problems using limited memory for two prototypical graph problems: maximum weighted matching and minimum weighted edge cover. For matching, we conduct a thorough computational study on two of the semi-streaming algorithms including a recent breakthrough result that achieves a $1/(2+\varepsilon)$-approximation of the weight while using $O( n \log W /\epsilon)$ memory (here $n$ is the number of vertices and $W$ is the maximum edge weight), designed by Paz and Schwartzman [SODA, 2017]. Empirically, we show that the semi-streaming algorithms produce matchings whose weight is close to the best $1/2$-approximate offline algorithm while requiring less time and an order-of-magnitude less memory. For minimum weighted edge cover, we develop three novel semi-streaming algorithms. Two of these algorithms require a single pass through the input graph, require $O(n \log n)$ memory, and provide a 2-approximation guarantee on the objective. We also leverage a relationship between approximate maximum weighted matching and approximate minimum weighted edge cover to develop a two-pass $3/2+\epsilon$-approximate algorithm with the memory requirement of Paz and Schwartzman's semi-streaming matching algorithm. These streaming approaches are compared against the state-of-the-art 3/2-approximate offline algorithm. The semi-streaming matching and the novel edge cover algorithms proposed in this paper can process graphs with several billions of edges in under 30 minutes using 6 GB of memory, which is at least an order of magnitude improvement from the offline (non-streaming) algorithms. For the largest graph, the best alternative offline parallel approximation algorithm (GPA+ROMA) could not finish in three hours even while employing hundreds of processors and 1 TB of memory. We also demonstrate an application of the semi-streaming algorithm by computing a matching using linearly bounded memory on item intersection graphs derived from three machine learning datasets, whereas the existing offline algorithms could not complete on one of these datasets since their memory requirements exceeded 1TB.
Population count is a primitive used in many applications. Commodity processors have dedicated instructions for achieving high-performance population count. Motivated by the productivity of high-level synthesis and the importance of population count, in this paper we investigated the OpenCL implementations of population count algorithms, and evaluated their performance and resource utilizations on an FPGA. Based on the results, we select the most efficient implementation. Then we derived a reduction pattern from a representative application of population count. We parallelized the reduction with atomic functions, and optimized it with vectorized memory accesses, tree reduction, and compute-unit duplication. We evaluated the performance of the reduction kernel on an Intel (R) Xeon (R) CPU and an Intel (R) IrisT Pro integrated GPU, and an FPGA card that features an Intel (R) Arria (R) 10 FPGA. When DRAM memory bandwidth is comparable on the three computing platforms, the FPGA can achieve the highest kernel performance for large workload. On the other hand, we described performance bottlenecks on the FPGA. To make FPGAs more competitive in raw performance compared to high-performant CPU and GPU platforms, it is important to increase external memory bandwidth, minimize data movement between a host and a device, and reduce OpenCL runtime overhead on an FPGA.
Mark III Hypercube concept applied in development of series of increasingly powerful computers. Processor of each node of Mark III Hypercube ensemble is specialized computer containing three subprocessors and shared main memory. Solves problem quickly by simultaneously processing part of problem at each such node and passing combined results to host computer. Disciplines benefitting from speed and memory capacity include astrophysics, geophysics, chemistry, weather, high-energy physics, applied mechanics, image processing, oil exploration, aircraft design, and microcircuit design.
Current methods for stochastic media transport are either computationally expensive or, by nature, approximate. Moreover, none of the well-developed, benchmarked approximate methods can compute the variance caused by the stochastic mixing, a quantity especially important to safety calculations. Therefore, we derive and apply a new conditional probability function (CPF) for use in the recently developed stochastic media transport algorithm Conditional Point Sampling (CoPS), which 1) leverages the full intra-particle memory of CoPS to yield errorless computation of stochastic media outputs in 1D, binary, Markovian-mixed media, and 2) leverages the full inter-particle memory of CoPS and the recently developed Embedded Variance Deconvolution method to yield computation of the variance in transport outputs caused by stochastic material mixing. Numerical results demonstrate errorless stochastic media transport as compared to reference benchmark solutions with the new CPF for this class of stochastic mixing as well as the ability to compute the variance caused by the stochastic mixing via CoPS. Using previously derived, non-errorless CPFs, CoPS is further found to be more accurate than the atomic mix approximation, Chord Length Sampling (CLS), and most of memory-enhanced versions of CLS surveyed. In addition, we study the compounding behavior of CPF error as a function of cohort size (where a cohort is a group of histories that share intra-particle memory) and recommend that small cohorts be used when computing the variance in transport outputs caused by stochastic mixing.
The Hubble Space Telescope (HST) team is preparing for NASA's third scheduled service call to Hubble. This mission, STS-103, will launch from Kennedy Space Center aboard the Space Shuttle Discovery. The seven flight crew members are Commander Curtis L. Brown, Pilot Scott J. Kelly, European Space Agency (ESA) astronaut Jean-Francois Clervoy who will join space walkers Steven L. Smith, C. Michael Foale, John M. Grunsfeld, and ESA astronaut Claude Nicollier. The objectives of the HST Third Servicing Mission (SM3A) are to replace the telescope's six gyroscopes, a Fine-Guidance Sensor, an S-Band Single Access Transmitter, a spare solid-state recorder and a high-voltage/temperature kit for protecting the batteries from overheating. In addition, the crew plans to install an advanced computer that is 20 times faster and has six times the memory of the current Hubble Space Telescope computer. To prepare for these extravehicular activities (EVAs), the SM3A astronauts participated in Crew Familiarization sessions with the actual SM3A flight hardware. During these sessions the crew spent long hours rehearsing their space walks in the Guidance Navigation Simulator and NBL (Neutral Buoyancy Laboratory). Using space gloves, flight Space Support Equipment (SSE), and Crew Aids and Tools (CATs), the astronauts trained with and verified flight orbital replacement unit (ORU) hardware. The crew worked with a number of trainers and simulators, such as the High Fidelity Mechanical Simulator, Guidance Navigation Simulator, System Engineering Simulator, the Aft Shroud Door Trainer, the Forward Shell/Light Shield Simulator, and the Support Systems Module Bay Doors Simulator. They also trained and verified the flight Orbital Replacement Unit Carrier (ORUC) and its ancillary hardware. Discovery's planned 10-day flight is scheduled to end with a night landing at Kennedy.
Here, we offer a modern interpretation of Lagrangian hydrodynamics as employed in Lagrangian simulations of compressible fluid flow. Our main result is to show that artificial viscosity, traditionally viewed as a numerical artifice to control unphysical oscillations in flows with shocks, actually represents a physical process and is necessary to derive accurate simulations in any compressible flow. We begin by reviewing the origins of two numerical devices, artificial viscosity and finite-volume methods. We proceed to construct a mathematical (PDE) model that incorporates those numerics and in which a new length scale, the observer, arises representing the discretization. Associated with that length scale, there are new inviscid fluxes that are the artificial viscosity as first formulated by Richtmyer and an artificial heat flux postulated by Noh but typically not included in Lagrangian codes. We discuss the connection of our results to bivelocity hydrodynamics. We conclude with some speculation as to the direction of future developments in multidimensional Lagrangian codes as computers get faster and have larger memories.
An algorithm for solving the discrete adjoint system based on an unstructured-grid discretization of the Navier-Stokes equations is presented. The method is constructed such that an adjoint solution exactly dual to a direct differentiation approach is recovered at each time step, yielding a convergence rate which is asymptotically equivalent to that of the primal system. The new approach is implemented within a three-dimensional unstructured-grid framework and results are presented for inviscid, laminar, and turbulent flows. Improvements to the baseline solution algorithm, such as line-implicit relaxation and a tight coupling of the turbulence model, are also presented. By storing nearest-neighbor terms in the residual computation, the dual scheme is computationally efficient, while requiring twice the memory of the flow solution. The scheme is expected to have a broad impact on computational problems related to design optimization as well as error estimation and grid adaptation efforts.
Molecular spintronics devices (MSDs) attempt to harness molecules’ quantum state, size, and configurable attributes for application in computer devices—a quest that began more than 70 years ago. In the vast number of theoretical studies and limited experimental attempts, MSDs have been found to be suitable for application in memory devices and futuristic quantum computers. MSDs have recently also exhibited intriguing spin photovoltaic-like phenomena, signaling their potential application in cost-effective and novel solar cell technologies. The molecular spintronics field’s major challenge is the lack of mass-fabrication methods producing robust magnetic molecule connections with magnetic electrodes of different anisotropies. Another main challenge is the limitations of conventional theoretical methods for understanding experimental results and designing new devices. Magnetic tunnel junction-based molecular spintronics devices (MTJMSDs) are designed by covalently connecting paramagnetic molecules across an insulating tunneling barrier. The insulating tunneling barrier serves as a mechanical spacer between two ferromagnetic (FM) electrodes of tailorable magnetic anisotropies to allow molecules to undergo many intriguing phenomena. Our experimental studies showed that the paramagnetic molecules could produce strong antiferromagnetic coupling between two FM electrodes, leading to a dramatic large-scale impact on the magnetic electrode itself. Recently, we showed that the Monte Carlo Simulation (MCS) was effective in providing plausible insights into the observation of unusual magnetic domains based on the role of single easy-axis magnetic anisotropy. Here, we experimentally show that the response of a paramagnetic molecule is dramatically different when connected to FM electrodes of different easy-axis anisotropies. Motivated by our experimental studies, here, we report on an MCS study investigating the impact of the simultaneous presence of two easy-axis anisotropies on MTJMSD equilibrium properties. In-plane easy-axis anisotropy produced multiple magnetic phases of opposite spins. The multiple magnetic phases vanished at higher thermal energy, but the MTJMSD still maintained a higher magnetic moment because of anisotropy. The out-of-plane easy-axis anisotropy caused a dominant magnetic phase in the FM electrode rather than multiple magnetic phases. The simultaneous application of equal-magnitude in-plane and out-of-plane easy-axis anisotropies on the same electrode negated the anisotropy effect. Our experimental and MCS study provides insights for designing and understanding new spintronics-based devices.
The explicit transient dynamics technology in use today for simulating the impact and subsequent transient dynamic response of a structure has its origins in the 'hydrocodes' dating back to the late 1940's. The growth in capability in explicit transient dynamics technology parallels the growth in speed and size of digital computers. Computer software for simulating the explicit transient dynamic response of a structure is characterized by algorithms that use a large number of small steps. In explicit transient dynamics software there is a significant emphasis on speed and simplicity. The finite element technology used to generate the spatial discretization of a structure is based on a compromise between completeness of the representation for the physical processes modelled and speed in execution. That is, since it is expected in every calculation that the deformation will be finite and the material will be strained beyond the elastic range, the geometry and the associated gradient operators must be reconstructed, as well as complex stress-strain models evaluated at every time step. As a result, finite elements derived for explicit transient dynamics software use the simplest and barest constructions possible for computational efficiency while retaining an essential representation of the physical behavior. The best example of this technology is the four-node bending quadrilateral derived by Belytschko, Lin and Tsay. Today, the speed, memory capacity and availability of computer hardware allows a number of the previously used algorithms to be 'improved.' That is, it is possible with today's computing hardware to modify many of the standard algorithms to improve their representation of the physical process at the expense of added complexity and computational effort. The purpose is to review a number of these algorithms and identify the improvements possible. In many instances, both the older, faster version of the algorithm and the improved and somewhat slower version of the algorithm are found implemented together in software. Specifically, the following seven algorithmic items are examined: the invariant time derivatives of stress used in material models expressed in rate form; incremental objectivity and strain used in the numerical integration of the material models; the use of one-point element integration versus mean quadrature; shell elements used to represent the behavior of thin structural components; beam elements based on stress-resultant plasticity versus cross-section integration; the fidelity of elastic-plastic material models in their representation of ductile metals; and the use of Courant subcycling to reduce computational effort.
In this work, we demonstrate SONOS (silicon-oxide-nitrideoxide- silicon) analog memory arrays that are optimized for neural network inference. The devices are fabricated in a 40nm process and operated in the subthreshold regime for in-memory matrix multiplication. Subthreshold operation enables low conductances to be implemented with low error, which matches the typical weight distribution of neural networks, which is heavily skewed toward near-zero values. This leads to high accuracy in the presence of programming errors and process variations. We simulate the end-to-end neural network inference accuracy, accounting for the measured programming error, read noise, and retention loss in a fabricated SONOS array. Evaluated on the ImageNet dataset using ResNet50, the accuracy using a SONOS system is within 2.16% of floating-point accuracy without any retraining. The unique error properties and high On/Off ratio of the SONOS device allow scaling to large arrays without bit slicing, and enable an inference architecture that achieves 20 TOPS/W on ResNet50, a >10× gain in energy efficiency over state-of-the-art digital and analog inference accelerators.
More reliable computers could be assembled by connecting four proposed VSLI "building block" circuits with built-in error detection to standard microprocessors and memory devices to form self checking computer module. Each building block detects its own malfunctions and single bit errors found in memory.
Efforts being expended at NASA-Langley to define a validation methodology, techniques for comparing advanced systems concepts, and design guidelines for characterizing fault tolerant digital avionics are described with an emphasis on the capabilities of AIRLAB, an environmentally controlled laboratory. AIRLAB has VAX 11/750 and 11/780 computers with an aggregate of 22 Mb memory and over 650 Mb storage, interconnected at 256 kbaud. An additional computer is programmed to emulate digital devices. Ongoing work is easily accessed at user stations by either chronological or key word indexing. The CARE III program aids in analyzing the capabilities of test systems to recover from faults. An additional code, the semi-Markov unreliability program (SURE) generates upper and lower reliability bounds. The AIRLAB facility is mainly dedicated to research on designs of digital flight-critical systems which must have acceptable reliability before incorporation into aircraft control systems. The digital systems would be too costly to submit to a full battery of flight tests and must be initially examined with the AIRLAB simulation capabilities.