Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “efficient exploration”

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 1,585 records · Page 88

Environmentally Robust Quantum Dot Down Converters for Highly Efficient Solid State Lighting

This project contributes fundamental understanding of the chemical synthesis of cadmium free quantum dots (QDs) designed for solid state lighting (SSL). Methods to incorporate gallium into quantum dot architectures were explored, which led to more stable and brighter solid state lighting devices. These findings will help advance the incorporation of quantum dots in highly energy efficient lighting devices with a warm color spectrum. Such Cd-free lighting devices can greatly accelerate the adoption of high efficiency, warm spectrum lighting in buildings. The project targeted InP/GaP/ZnS colloidal quantum dots that more effectively absorb the blue light produced by light emitting diode packages and emit monochromatic red light with near perfect efficiency. While significant improvements in the QD absorptivity were demonstrated, the absorbance increase achieved was 3x rather than the targeted 10x increase. Similarly, the luminescence from these QDs achieved about half of the target photoluminescence quantum yield (PLQY) and twice the desired spectral width, a measure of the color purity of their luminescence. While the target characteristics and some of the year 2/2 milestones were not met, the new QD architectures did lead to greater stability on SSL packages, a sign of the long term promise of the technology.

36 MATERIALS SCIENCE↗

Employing a Narrow-Band-Gap Mediator in Ternary Solar Cells for Enhanced Photovoltaic Performance

Ternary organic solar cells (OSCs) provide a convenient and effective means to further improve the power conversion efficiency (PCE) of binary ones via composition control. However, the role of the third component remains to be explored in specific binary systems. Herein, we report ternary blend solar cells by adding the narrow-band-gap donor PCE10 as the mediator into the PBDB-T:IDTT-T binary blend system. The extended absorption, efficient fluorescence resonance energy transfer, enhanced charge dissociation, and induced tighter molecular packing of the ternary blend films enhance the photovoltaic properties of devices and deliver a champion PCE of 10.73% with an impressively high open-circuit voltage (VOC) of 1.03 V. Good miscibility and similar molecular packing behavior of the components guarantee the desired morphology in the ternary blend films, leading to solar cell devices with over 10% PCEs at a range of compositions. Our results suggest that ternary systems with properly aligned energy levels and overlapping absorption among the components hold great promises to further enhance the performance of corresponding binary ones.

14 SOLAR ENERGY↗

Automated Unstructured Grid Adaptation on a Strut Fuel Injector at Hypervelocity Flow Conditions

Computational fluid dynamics (CFD) analysis is presented with the use of an automated unstructured grid adaptation tool on a strut fuel injector at hypervelocity flow conditions. The analysis was carried out with the VULCAN-CFD solver using Reynolds-averaged simulations (RAS). The hypervelocity flow conditions match the high Mach number flow of the experiments conducted as part of the Enhanced Injection and Mixing Project (EIMP) at the NASA Langley Research Center (LaRC). The current work uses an automated grid adaptation tool recently implemented in VULCAN-CFD, and explores this tool’s ability to solve highspeed mixing problems. Simulation results obtained using the unstructured adaptive grid approach are compared to those on a user-generated structured grid. These results are evaluated by analyzing how efficiently comparable fidelity results are obtained from both adapted and structured simulations. In addition, two adaptation strategies were used to explore the impact on the final solution. In the current work, the unstructured grid adaptation tool automatically generates unstructured grids and performs adaptation of the grid based on a Hessian error estimate of a specified flowfield parameter. Multiple adaptations were executed using each run strategy with the one-dimensional values of the mixing efficiency used to determine grid convergence and for comparison with the structured grid simulation results. It was found that the unstructured adaptive grid simulations were able to produce results that matched closely with those on structured grids using far fewer grid cells, and thus, requiring far less computational time to reach the solution. It was also discovered that the adaptation run strategy influenced the total number of grid cells and the efficiency with which a final grid-adapted solution was reached. Furthermore, motivated by the grid convergence index (GCI) used for structured grid simulations, a grid convergence estimate (GCE) was developed and demonstrated for the grid adaptation. Overall, the investigation demonstrated that the automated unstructured grid adaptation tool implemented in VULCAN-CFD is capable of accurately and efficiently solving complex high-speed mixing problems with only a fraction of the grid cells required to obtain comparable results on a user-generated structured grid.

hypersonics↗

Automated Unstructured Grid Adaptation on a Strut Fuel Injector at Hypervelocity Flow Conditions

Computational fluid dynamics (CFD) analysis is presented with the use of an automated unstructured grid adaptation tool on a strut fuel injector at hypervelocity flow conditions. The analysis was carried out with the VULCAN-CFD solver using Reynolds-averaged simulations (RAS). The hypervelocity flow conditions match the high Mach number flow of the experiments conducted as part of the Enhanced Injection and Mixing Project (EIMP) at the NASA Langley Research Center (LaRC). The current work uses an automated grid adaptation tool recently implemented in VULCAN-CFD, and explores this tool’s ability to solve highspeed mixing problems. Simulation results obtained using the unstructured adaptive grid approach are compared to those on a user-generated structured grid. These results are evaluated by analyzing how efficiently comparable fidelity results are obtained from both adapted and structured simulations. In addition, two adaptation strategies were used to explore the impact on the final solution. In the current work, the unstructured grid adaptation tool automatically generates unstructured grids and performs adaptation of the grid based on a Hessian error estimate of a specified flowfield parameter. Multiple adaptations were executed using each run strategy with the one-dimensional values of the mixing efficiency used to determine grid convergence and for comparison with the structured grid simulation results. It was found that the unstructured adaptive grid simulations were able to produce results that matched closely with those on structured grids using far fewer grid cells, and thus, requiring far less computational time to reach the solution. It was also discovered that the adaptation run strategy influenced the total number of grid cells and the efficiency with which a final grid-adapted solution was reached. Furthermore, motivated by the grid convergence index (GCI) used for structured grid simulations, a grid convergence estimate (GCE) was developed and demonstrated for the grid adaptation. Overall, the investigation demonstrated that the automated unstructured grid adaptation tool implemented in VULCAN-CFD is capable of accurately and efficiently solving complex high-speed mixing problems with only a fraction of the grid cells required to obtain comparable results on a user-generated structured grid.

hypersonics↗

Solution of the Schrödinger equation for quasi-one-dimensional materials using helical waves

We formulate and implement a spectral method for solving the Schrödinger equation, as it applies to quasi-one-dimensional materials and structures. This allows for computation of the electronic structure of important technological materials such as nanotubes (of arbitrary chirality), nanowires, nanoribbons, chiral nanoassemblies, nanosprings and nanocoils, in an accurate, efficient and systematic manner. Our work is motivated by the observation that one of the most successful methods for carrying out electronic structure calculations of bulk/crystalline systems — the plane-wave method — is a spectral method based on eigenfunction expansion. Our scheme avoids computationally onerous approximations involving periodic supercells often employed in conventional plane-wave calculations of quasi-one-dimensional materials, and also overcomes several limitations of other discretization strategies, e.g., those based on finite differences and atomic orbitals. The basis functions in our method — called helical waves (or twisted waves) — are eigenfunctions of the Laplacian with symmetry adapted boundary conditions, and are expressible in terms of plane waves and Bessel functions in helical coordinates. We describe the setup of fast transforms to carry out discretization of the governing equations using our basis set, and the use of matrix-free iterative diagonalization to obtain the electronic eigenstates. Miscellaneous computational details, including the choice of eigensolvers, use of a preconditioning scheme, evaluation of oscillatory radial integrals and the imposition of a kinetic energy cutoff are discussed. We have implemented these strategies into a computational package called HelicES (Helical Electronic Structure). We demonstrate the utility of our method in carrying out systematic electronic structure calculations of various quasi-one-dimensional materials through numerous examples involving nanotubes, nanoribbons and nanowires. We also explore the convergence properties of our method, and assess its accuracy and computational efficiency by comparison against reference finite difference, transfer matrix method and plane-wave results. We anticipate that our method will find applications in computational nanomechanics and multiscale modeling, for carrying out transport calculations of interest to the field of semiconductor devices, and for the discovery of novel chiral phases of matter that are of relevance to the burgeoning quantum hardware industry.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

An adaptive Kriging surrogate method for efficient uncertainty quantification with an application to geological carbon sequestration modeling

We present that in numerical modeling of geological carbon sequestration (GCS), uncertainty quantification (UQ) is usually needed to evaluate the impact of uncertain model parameters on model predictions caused by limited measurements and incomplete knowledge of the parameters. However, UQ for GCS is computationally expensive due to the large ensemble of complex and lengthy model simulations. In this study, we propose an adaptive Kriging method to build a fast-to-evaluate surrogate of the GCS model to alleviate the heavy computational burden. The surrogate model is efficiently generated using a Taylor expansion-based adaptive experimental design algorithm that combines a distance-based exploration criterion and an exploitation criterion to adaptively search for informative training samples. In addition, we analyze the uncertainty brought by substituting the surrogate for the actual simulation model and explore its influence on UQ results. Lastly, our method is demonstrated in a synthetic GCS model and its performance is evaluated in comparison with the conventional Monte Carlo sampling. Results indicate that our method can greatly improve the computational efficiency in UQ and provide an effective and reliable UQ solution with the consideration of surrogate uncertainty.

58 GEOSCIENCES↗

First-Principles-Based Machine Learning Models for Phase Behavior and Transport Properties of CO 2

In this work, we construct distinct first-principles-based machine-learning models of CO 2 , reproducing the potential energy surface of the PBE-D3, BLYP-D3, SCAN and SCAN-rvv10 approximations of density functional theory. Here we employ the Deep Potential methodology to develop the models and consequently achieve a significant computational efficiency over ab initio molecular dynamics (AIMD) that allows for larger system sizes and time scales to be explored. Although our models are trained only with liquid phase configurations, they are able to simulate a stable interfacial system and predict vapor-liquid equilibrium properties, in good agreement with results from the literature. Because of the computational efficiency of the models, we are also able to obtain transport properties, such as viscosity and diffusion coefficients. We find that the SCAN-based model presents a temperature shift in the position of the critical point, while the SCAN-rvv10-based model shows improvement but still exhibits a temperature shift that remains approximately constant for all properties investigated in this work. We find that the BLYP-D3-based model generally performs better for liquid phase and vapor-liquid equilibrium properties, but the PBE-D3-based model is better suited for predicting transport properties.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Alternative Formats to Achieve More Efficient Energy Codes for Commercial Buildings

This paper identifies and examines several formats or structures that could be used to create the next generation of more efficient energy codes and standards for commercial buildings. Pacific Northwest National Laboratory (PNNL) is funded by the U.S. Department of Energy’s Building Energy Codes Program (BECP) to provide technical support to the development of ANSI/ASHRAE/IES Standard 90.1. While the majority of PNNL’s ASHRAE Standard 90.1 support focuses on developing and evaluating new requirements, a portion of its work involves consideration of the format of energy standards. In its current working plan, the ASHRAE 90.1 committee has approved an energy goal of 50% improvement in Standard 90.1-2013 relative to Standard 90.1-2004, and will likely be considering higher improvement targets for future versions of the standard. To cost-effectively achieve the 50% goal in manner that can gain stakeholder consensus, formats other than prescriptive must be considered. Alternative formats that include reducing the reliance on prescriptive requirements may make it easier to achieve these aggressive efficiency levels in new codes and standards. The focus on energy code and standard formats is meant to explore approaches to presenting the criteria that will foster compliance, enhance verification, and stimulate innovation while saving energy in buildings. New formats may also make it easier for building designers and owners to design and build the levels of efficiency called for in the new codes and standards. This paper examines a number of potential formats and structures, including prescriptive, performance-based (with sub-formats of performance equivalency and performance targets), capacity constraint-based, and outcome-based. The paper also discusses the pros and cons of each format from the viewpoint of code users and of code enforcers.

Conover, David R.↗

Spin-neurons: A possible path to energy-efficient neuromorphic computers

Recent years have witnessed growing interest in the field of brain-inspired computing based on neural-network architectures. In order to translate the related algorithmic models into powerful, yet energy-efficient cognitive-computing hardware, computing-devices beyond CMOS may need to be explored. The suitability of such devices to this field of computing would strongly depend upon how closely their physical characteristics match with the essential computing primitives employed in such models. In this work, we discuss the rationale of applying emerging spin-torque devices for bio-inspired computing. Recent spin-torque experiments have shown the path to low-current, low-voltage, and high-speed magnetization switching in nano-scale magnetic devices. Such magneto-metallic, current-mode spin-torque switches can mimic the analog summing and “thresholding” operation of an artificial neuron with high energy-efficiency. Comparison with CMOS-based analog circuit-model of a neuron shows that “spin-neurons” (spin based circuit model of neurons) can achieve more than two orders of magnitude lower energy and beyond three orders of magnitude reduction in energy-delay product. The application of spin-neurons can therefore be an attractive option for neuromorphic computers of future.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Direct Observation of Structural Evolution of Metal Chalcogenide in Electrocatalytic Water Oxidation

As one of the most remarkable oxygen evolution reaction (OER) electrocatalysts, metal chalcogenides have been intensively reported during the past few decades because of their high OER activities. It has been reported that electron-chemical conversion of metal chalcogenides into oxides/hydroxides would take place after the OER. However, the transition mechanism of such unstable structures, as well as the real active sites and catalytic activity during the OER for these electrocatalysts, has not been understood yet; therefore a direct observation for the electrocatalytic water oxidation process, especially at nano or even angstrom scale, is urgently needed. In this work, by employing advanced Cs-corrected transmission electron microscopy (TEM), a step by step oxidational evolution of amorphous electrocatalyst CoS x into crystallized CoOOH in the OER has been in situ captured: irreversible conversion of CoS x to crystallized CoOOH is initiated on the surface of the electrocatalysts with a morphology change via Co(OH) 2 intermediate during the OER measurement, where CoOOH is confirmed as the real active species. Besides, this transition process has also been confirmed by multiple applications of X-ray photoelectron spectroscopy (XPS), in situ Fourier-transform infrared spectroscopy (FTIR), and other ex situ technologies. Moreover, on the basis of this discovery, a high-efficiency electrocatalyst of a nitrogen-doped graphene foam (NGF) coated by CoS x has been explored through a thorough structure transformation of CoOOH. In conclusion, we believe this in situ and in-depth observation of structural evolution in the OER measurement can provide insights into the fundamental understanding of the mechanism for the OER catalysts, thus enabling the more rational design of low-cost and high-efficient electrocatalysts for water splitting.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

The Asteroid Redirect Mission (ARM): Exploration of a Former Binary NEA?

The National Aeronautics and Space Administration (NASA) is developing the Asteroid Redirect Mission (ARM) as a capability demonstration for future human exploration, including use of high-power solar electric propulsion, which allows for the efficient movement of large masses through deep space. The ARM will also demonstrate the capability to conduct proximity operations with natural space objects and crewed operations beyond the security of quick Earth return. The Asteroid Redirect Robotic Mission (ARRM), currently in formulation, will visit a large near-Earth asteroid (NEA), collect a multi-ton boulder from its surface, conduct a demonstration of a slow push planetary defense technique, and redirect the multi-ton boulder into a stable orbit around the Moon. Once returned to cislunar space in the mid-2020s, astronauts aboard an Orion spacecraft will dock with the robotic vehicle to explore the boulder and return samples to Earth. The ARM is part of NASA's plan to advance technologies, capabilities, and spaceflight experience needed for a human mission to the Martian system in the 2030s. The ARM and subsequent availability of the asteroidal material in cis-lunar space, provide significant opportunities to advance our knowledge of small bodies in the synergistic areas of science, planetary defense, and in-situ resource utilization (ISRU). The current reference target for the ARM is NEA (341843) 2008 EV5, which may have been the primary body of a former binary system (Busch et al., 2011; Tardivel et al., 2016). The ARRM will perform several close proximity operations to investigate the NEA and map its surface. A detailed investigation of this object may allow a better understanding of binary NEA physical characteristics and the possible outcomes for their evolution. An overview of the ARM robotic and crewed segments, including mission operations, and a discussion of potential opportunities for participation with the ARM will be provided in this presentation.

Abell, P. A.↗

An Overview of NASA's Asteroid Redirect Mission (ARM) Concept

The National Aeronautics and Space Administration (NASA) is developing the Asteroid Redirect Mission (ARM) as a capability demonstration for future human exploration, including use of high-power solar electric propulsion, which allows for the efficient movement of large masses through deep space. The ARM will also demonstrate the capability to conduct proximity operations with natural space objects and crewed operations beyond the security of quick Earth return. The Asteroid Redirect Robotic Mission (ARRM), currently in formulation, will visit a large near-Earth asteroid (NEA), collect a multi-ton boulder from its surface, conduct a demonstration of a slow push planetary defense technique, and redirect the multi-ton boulder into a stable orbit around the Moon. Once returned to cislunar space in the mid-2020s, astronauts aboard an Orion spacecraft will dock with the robotic vehicle to explore the boulder and return samples to Earth. The ARM is part of NASA's plan to advance technologies, capabilities, and spaceflight experience needed for a human mission to the Martian system in the 2030s. The ARM and subsequent availability of the asteroidal material in cis-lunar space, provide significant opportunities to advance our knowledge of small bodies in the synergistic areas of science, planetary defense, and in-situ resource utilization (ISRU). NASA established the Formulation Assessment and Support Team (FAST), comprised of scientists, engineers, and technologists, which supported ARRM mission requirements formulation, answered specific questions concerning potential target asteroid physical properties, and produced a publically available report. The ARM Investigation Team is being organized to support ARM implementation and execution. NASA is also open to collaboration with its international partners and welcomes further discussions. An overview of the ARM robotic and crewed segments, including mission requirements, NEA targets, and mission operations, and a discussion of potential opportunities for participation with the ARM will be provided.

Abell, P. A.↗

Automated Unstructured Grid Adaptation on a Strut Fuel Injector at Hypervelocity Flow Conditions

Computational fluid dynamics (CFD) analysis is presented with the use of an automated unstructured grid adaptation tool on a strut fuel injector at hypervelocity flow conditions. The analysis was carried out with the VULCAN-CFD solver using Reynolds-averaged simulations (RAS). The hypervelocity flow conditions match the high Mach number flow of the experiments conducted as part of the Enhanced Injection and Mixing Project (EIMP) at the NASA Langley Research Center (LaRC). The current work utilizes an automated grid adaptation tool recently implemented into VULCAN-CFD, and explores this tool’s ability to solve high-speed mixing problems. Simulation results obtained using the unstructured adaptive grid approach are compared to those on a user generated structured grid. These results are evaluated by analyzing how efficiently comparable fidelity results are obtained from both adapted and structured simulations. In addition, two adaptation strategies were used to explore the impact on the final solution. In the current work, the unstructured grid adaptation tool automatically generates unstructured grids and performs adaptation of the grid based on a Hessian error estimate of a specified flow field parameter. Multiple adaptations were executed using each run strategy with the one-dimensional values of the mixing efficiency used to determine grid convergence and for comparison with the structured grid simulation results. It was found that the unstructured adaptive grid simulations were able to produce results that matched closely with those on structured grids using far fewer grid cells, and thus, requiring far less computational time to reach the solution. It was also discovered that the adaptation run strategy influenced the total number of grid cells and the efficiency with which a final grid-adapted solution was reached. Overall, the investigation demonstrated that the automated unstructured grid adaptation tool implemented in VULCAN-CFD is capable of accurately and efficiently solving complex highspeed mixing problems using only a fraction of the grid cells required to obtain comparable results using a user-generated structured grid.

grid adaptation↗

Exploring magnetized liner inertial fusion with a semi-analytic model

In this paper, we explore magnetized liner inertial fusion (MagLIF) [S. A. Slutz et al., Phys. Plasmas 17, 056303 (2010)] using a semi-analytic model [R. D. McBride and S. A. Slutz, Phys. Plasmas 22, 052708 (2015)]. Specifically, we present simulation results from this model that: (a) illustrate the parameter space, energetics, and overall system efficiencies of MagLIF; (b) demonstrate the dependence of radiative loss rates on the radial fraction of the fuel that is preheated; (c) explore some of the recent experimental results of the MagLIF program at Sandia National Laboratories [M. R. Gomez et al., Phys. Rev. Lett. 113, 155003 (2014)]; (d) highlight the experimental challenges presently facing the MagLIF program; and (e) demonstrate how increases to the preheat energy, fuel density, axial magnetic field, and drive current could affect future MagLIF performance.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Nanocrystals-Related Synthesis, Assembly, and Energy Applications

During the past decades, nanocrystals have attracted broad attention due to their unique shape- and size-dependent physical and chemical properties that differ drastically from their bulk counterparts. Hitherto, much effort has been dedicated to achieving rational controlling over the morphology, assembly, and related energy applications of the nanomaterials. Therefore, the ability to manipulate the morphology, size, and size distribution of inorganic nanomaterials is still an important goal in modern materials physics and chemistry. Especially, the world s demand for energy supply is causing a dramatic escalation of social and political unrest. Likewise, the environmental impact of the global climate change due to the combustion of fossil fuel is becoming increasingly alarming. These problems compel us to search for effective routes to build devices that can supply sustainable energy, with not only high efficiency but also environmental friendship. One of ways to relieve the energy crisis is to exploit devices based on renewable energy sources, such as solar energy and water power. Aiming at this exploration, the primary stage requires the design of appropriate strategies for the synthesis of high-quality nanocrystals with respect to size uniformity and superior electrochemical performances. As a consequence, we organize the current special issue for Journal of Nanomaterials to provide the authors with a platform and readers with the latest achievements of nanocrystals-related synthesis, assembly, and energy applications.

14 SOLAR ENERGY↗

International Space Station (ISS) Bacterial Filter Elements (BFEs): Filter Efficiency and Pressure Drop Testing of Returned Units

The air quality control equipment aboard the International Space Station (ISS) and future deep space exploration vehicles provide the vital function of maintaining a clean cabin environment for the crew and the hardware. This becomes a serious challenge in pressurized space compartments since no outside air ventilation is possible, and a larger particulate load is imposed on the filtration system due to lack of sedimentation. The ISS Environmental Control and Life Support (ECLS) system architecture in the U.S. Segment uses a distributed particulate filtration approach consisting of traditional High-Efficiency Particulate Air (HEPA) filters deployed at multiple locations in each U.S. Seg-ment module; these filters are referred to as Bacterial Filter Elements, or BFEs. In our previous work, we presented results of efficiency and pressure drop measurements for a sample set of two returned BFEs with a service life of 2.5 years. In this follow-on work, we present similar efficiency, pressure drop, and leak tests results for a larger sample set of six returned BFEs. The results of this work can aid the ISS Program in managing BFE logistics inventory through the stations planned lifetime as well as provide insight for managing filter element logistics for future exploration missions. These results also can provide meaningful guidance for particulate filter designs under consideration for future deep space exploration missions.

ISS Air Quality↗

Solid, structured composite neutron detectors with high dynamic range capability

Neutron detectors are essential across disciplines such as fundamental science, nuclear security, safeguards, and civilian applications. While 3 He-filled gas proportional counters have long been revered for their efficacy in detecting thermal neutrons and praised for their efficiency, neutron/gamma discrimination, and stability, the scarcity of 3He has spurred a search for alternatives. Here, we explore a solid structured scintillating particle composite (SPC) consisting of 6 Li-containing scintillating glass particles within an acrylic matrix as a neutron detector for high dynamic range applications. We show for the first time that an SPC neutron detector can boast an intrinsic detection efficiency of 0.261% for pure 252 Cf fission neutrons and an overall neutron detection efficiency of (0.546 ± 0.003)% at the Neutron Free-in-Air facility while being able to function in an intense gamma-ray environment. We also show that the SPC neutron detector supports fast neutron capture times and enables a dual-readout scheme that extends the detector dynamic range to high incident neutron fluxes. A scalable fabrication process allows for tailoring the SPC detector properties to the requirements of specific applications. Good agreement is found between the experimental results taken with a National Institute of Standards and Technology traceable 252 Cf source and the coupled MCNP6 and optical-ray-tracing simulations.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Improving Memory Subsystem Performance Using ViVA: Virtual Vector Architecture

The disparity between microprocessor clock frequencies and memory latency is a primary reason why many demanding applications run well below peak achievable performance. Software controlled scratchpad memories, such as the Cell local store, attempt to ameliorate this discrepancy by enabling precise control over memory movement; however, scratchpad technology confronts the programmer and compiler with an unfamiliar and difficult programming model. In this work, we present the Virtual Vector Architecture (ViVA), which combines the memory semantics of vector computers with a software-controlled scratchpad memory in order to provide a more effective and practical approach to latency hiding. ViVA requires minimal changes to the core design and could thus be easily integrated with conventional processor cores. To validate our approach, we implemented ViVA on the Mambo cycle-accurate full system simulator, which was carefully calibrated to match the performance on our underlying PowerPC Apple G5 architecture. Results show that ViVA is able to deliver significant performance benefits over scalar techniques for a variety of memory access patterns as well as two important memory-bound compact kernels, corner turn and sparse matrix-vector multiplication -- achieving 2x-13x improvement compared the scalar version. Overall, our preliminary ViVA exploration points to a promising approach for improving application performance on leading microprocessors with minimal design and complexity costs, in a power efficient manner.

97 MATHEMATICS AND COMPUTING↗