Search NASA⌕ Search

SEARCH · Search NASA

Results for “Performance Portability”

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 1,117 records · Page 62

Agile Multi-Scale Decompositions for Automatic Image Registration

In recent works, the first and third authors developed an automatic image registration algorithm based on a multiscale hybrid image decomposition with anisotropic shearlets and isotropic wavelets. This prototype showed strong performance, improving robustness over registration with wavelets alone. However, this method imposed a strict hierarchy on the order in which shearlet and wavelet features were used in the registration process, and also involved an unintegrated mixture of MATLAB and C code. In this paper, we introduce a more agile model for generating features, in which a flexible and user-guided mix of shearlet and wavelet features are computed. Compared to the previous prototype, this method introduces a flexibility to the order in which shearlet and wavelet features are used in the registration process. Moreover, the present algorithm is now fully coded in C, making it more efficient and portable than the MATLAB and C prototype. We demonstrate the versatility and computational efficiency of this approach by performing registration experiments with the fully-integrated C algorithm. In particular, meaningful timing studies can now be performed, to give a concrete analysis of the computational costs of the flexible feature extraction. Examples of synthetically warped and real multi-modal images are analyzed.

Science Data Processing↗

Agile Multi-Scale Decompositions for Automatic Image Registration

In recent works, the first and third authors developed an automatic image registration algorithm based on a multiscale hybrid image decomposition with anisotropic shearlets and isotropic wavelets. This prototype showed strong performance, improving robustness over registration with wavelets alone. However, this method imposed a strict hierarchy on the order in which shearlet and wavelet features were used in the registration process, and also involved an unintegrated mixture of MATLAB and C code. In this paper, we introduce a more agile model for generating features, in which a flexible and user-guided mix of shearlet and wavelet features are computed. Compared to the previous prototype, this method introduces a flexibility to the order in which shearlet and wavelet features are used in the registration process. Moreover, the present algorithm is now fully coded in C, making it more efficient and portable than the MATLAB and C prototype. We demonstrate the versatility and computational efficiency of this approach by performing registration experiments with the fully-integrated C algorithm. In particular, meaningful timing studies can now be performed, to give a concrete analysis of the computational costs of the flexible feature extraction. Examples of synthetically warped and real multi-modal images are analyzed.

Science Data Processing↗

Simulated X-Ray Radiographic Performance of a Bismuth-Loaded PVT Array

Recent material advancements in organic plastic scintillators enable marked increases in material detection efficiency, light yield, and pulse shape discrimination properties. These advances may resolve significant capability gaps for lowcost, portable, and durable dual-particle imaging (DPI) systems for nuclear safety, security, and safeguard purposes. One such material, a 21% bismuth-loaded polyvinyl toluene (BiPVT), is computationally evaluated as a small, pixelated radiographic array using Monte Carlo N-Particle (MCNP) and Zemax OpticStudio, and it is compared to identical evaluations of EJ-200 and EJ-256 arrays. MCNP software enables estimates of particle interaction and energy deposition, while OpticStudio computes optical light transport within each material. Here, computational estimates of spatial resolution and relative light collection at 370 kVp are found to agree with experimental results for both EJ-200 and EJ-256 arrays, thereby validating predictions of the same for the BiPVT array. As such, for equivalent exposures at 370 kVp, a BiPVT array may provide ~20× the light collection expected from EJ-200 and ~10× that expected from EJ-256. Similar comparisons of estimated light collection are also computed at 150 and 270 kVp, and these results suggest that BiPVT will provide significantly improved performance over EJ-200 and EJ-256 across all energies practical for portable X-ray radiography.

42 ENGINEERING↗

Small Portable PEM Fuel Cell Systems for NASA Exploration Missions

Oxygen-Hydrogen PEM-based fuel cell systems are being examined as a portable power source alternative in addition to advanced battery technology. Fuel cell power systems have been used by the Gemini, Apollo, and Space Shuttle programs. These systems have not been portable, but have been integral parts of their spacecraft, and have used reactants from a separate cryogenic supply. These systems typically have been higher in power. They also have had significant ancillary equipment sections that perform the pumping of reactants and coolant through the fuel cell stack and the separation of the product water from the unused reactant streams. The design of small portable fuel cell systems will be a significant departure from these previous designs. These smaller designs will have very limited ancillary equipment, relying on passive techniques for reactant and thermal management, and the reactant storage will be an integral part of the fuel cell system. An analysis of the mass and volume for small portable fuel cell systems was done to evaluate and quantify areas of technological improvement. A review of current fuel cell technology as well as reactant storage and management technology was completed to validate the analysis and to identify technology challenges

Burke, Kenneth A.↗

Challenge-driven printing strategies toward high-performance solid-state lithium batteries

Solid-state lithium batteries (SSLBs) are promising candidates for replacing traditional liquid-based Li-ion batteries and revolutionizing battery systems for electric vehicles and portable devices. However, longstanding issues such as form factors, interfacial contact resistance, balance between ion conductivity and mechanical strength, and manufacturing processability limit their applications. In this review we present how advanced printing technologies can help to mitigate typical problems in main components of SSLBs and improve device performance. Here we first introduce the common printing techniques for energy storage devices, then focus on the issues and corresponding printing strategies for anodes, cathodes, and solid-state electrolytes to guide the construction of energy-dense, free-form SSLBs. The features and effects of the printed structures are emphasized, as well. We conclude by discussing the problems associated with printing technologies and the potential research directions for printed solid-state batteries.

25 ENERGY STORAGE↗

Level-2 Milestone 5588: Deliver Strategic Plan and Initial Scalability Assessment by Advanced Architecture and Portability Specialists Team

This report documents the fact that the work in creating a strategic plan and beginning customer engagements has been completed. The description of milestone is: The newly formed advanced architecture and portability specialists (AAPS) team will develop a strategic plan to meet the goals of 1) sharing knowledge and experience with code teams to ensure that ASC codes run well on new architectures, and 2) supplying skilled computational scientists to put the strategy into practice. The plan will be delivered to ASC management in the first quarter. By the fourth quarter, the team will identify their first customers within PEM and IC, perform an initial assessment and scalability and performance bottleneck for next-generation architectures, and embed AAPS team members with customer code teams to assist with initial portability development within standalone kernels or proxy applications.

97 MATHEMATICS AND COMPUTING↗

An efficient and portable SIMD algorithm for charge/current deposition in Particle-In-Cell codes

In current computer architectures, data movement (from die to network) is by far the most energy consuming part of an algorithm (≈20pJ/word on-die to ≈10,000 pJ/word on the network). To increase memory locality at the hardware level and reduce energy consumption related to data movement, future exascale computers tend to use many-core processors on each compute nodes that will have a reduced clock speed to allow for efficient cooling. To compensate for frequency decrease, machine vendors are making use of long SIMD instruction registers that are able to process multiple data with one arithmetic operator in one clock cycle. SIMD register length is expected to double every four years. As a consequence, Particle-In-Cell (PIC) codes will have to achieve good vectorization to fully take advantage of these upcoming architectures. In this paper, we present a new algorithm that allows for efficient and portable SIMD vectorization of current/charge deposition routines that are, along with the field gathering routines, among the most time consuming parts of the PIC algorithm. Our new algorithm uses a particular data structure that takes into account memory alignment constraints and avoids gather/scat;ter instructions that can significantly affect vectorization performances on current CPUs. The new algorithm was successfully implemented in the 3D skeleton PIC code PICSAR and tested on Haswell Xeon processors (AVX2-256 bits wide data registers). Results show a factor of ×2 to ×2.5 speed-up in double precision for particle shape factor of orders 1–3. The new algorithm can be applied as is on future KNL (Knights Landing) architectures that will include AVX-512 instruction sets with 512 bits register lengths (8 doubles/16 singles). Program summary Program Title: vec_deposition Program Files doi:http://dx.doi.org/10.17632/nh77fv9k8c.1 Licensing provisions: BSD 3-Clause Programming language: Fortran 90 External routines/libraries: OpenMP > 4.0 Nature of problem: Exascale architectures will have many-core processors per node with long vector data registers capable of performing one single instruction on multiple data during one clock cycle. Data register lengths are expected to double every four years and this pushes for new portable solutions for efficiently vectorizing Particle-In-Cell codes on these future many-core architectures. One of the main hotspot routines of the PIC algorithm is the current/charge deposition for which there is no efficient and portable vector algorithm. Solution method: Here we provide an efficient and portable vector algorithm of current/charge deposition routines that uses a new data structure, which significantly reduces gather/scatter operations. Vectorization is controlled using OpenMP 4.0 compiler directives for vectorization which ensures portability across different architectures. Restrictions: Here we do not provide the full PIC algorithm with an executable but only vector routines for current/charge deposition. These scalar/vector routines can be used as library routines in your 3D Particle-In-Cell code. However, to get the best performances out of vector routines you have to satisfy the two following requirements: (1) Your code should implement particle tiling (as explained in the manuscript) to allow for maximized cache reuse and reduce memory accesses that can hinder vector performances. The routines can be used directly on each particle tile. (2) You should compile your code with a Fortran 90 compiler (e.g Intel, gnu or cray) and provide proper alignment flags and compiler alignment directives (more details in README file).

97 MATHEMATICS AND COMPUTING↗

DeepHyper: Asynchronous Hyperparameter Search for Deep Neural Networks

Hyperparameters employed by deep learning (DL) methods play a substantial role in the performance and reliability of these methods in practice. Unfortunately, finding performance-optimizing hyperparameter settings is a notoriously difficult task. Hyperparameter search methods typically have limited production-strength implementations or do not target scalability within a highly parallel machine, portability across different machines, experimental comparison between different methods, and tighter integration with workflow systems. In this paper, we present DeepHyper, a Python package that provides a common interface for the implementation and study of scalable hyperparameter search methods. It adopts the Balsam workflow system to hide the complexities of running large numbers of hyperparameter configurations in parallel on high-performance computing (HPC) systems. We implement and study asynchronous model-based search methods that consist of sampling a small number of input hyperparameter configurations and progressively fitting surrogate models over the input-output space until exhausting a user-defined budget of evaluations. We evaluate the efficacy of these methods relative to approaches such as random search, genetic algorithms, Bayesian optimization, and hyperband on DL benchmarks on CPU- and GPU-based HPC systems.

Balaprakash, Prasanna↗

The medical science DMZ: a network design pattern for data-intensive medical science

Abstract Objective We describe a detailed solution for maintaining high-capacity, data-intensive network flows (eg, 10, 40, 100 Gbps+) in a scientific, medical context while still adhering to security and privacy laws and regulations. Materials and Methods High-end networking, packet-filter firewalls, network intrusion-detection systems. Results We describe a “Medical Science DMZ” concept as an option for secure, high-volume transport of large, sensitive datasets between research institutions over national research networks, and give 3 detailed descriptions of implemented Medical Science DMZs. Discussion The exponentially increasing amounts of “omics” data, high-quality imaging, and other rapidly growing clinical datasets have resulted in the rise of biomedical research “Big Data.” The storage, analysis, and network resources required to process these data and integrate them into patient diagnoses and treatments have grown to scales that strain the capabilities of academic health centers. Some data are not generated locally and cannot be sustained locally, and shared data repositories such as those provided by the National Library of Medicine, the National Cancer Institute, and international partners such as the European Bioinformatics Institute are rapidly growing. The ability to store and compute using these data must therefore be addressed by a combination of local, national, and industry resources that exchange large datasets. Maintaining data-intensive flows that comply with the Health Insurance Portability and Accountability Act (HIPAA) and other regulations presents a new challenge for biomedical research. We describe a strategy that marries performance and security by borrowing from and redefining the concept of a Science DMZ, a framework that is used in physical sciences and engineering research to manage high-capacity data flows. Conclusion By implementing a Medical Science DMZ architecture, biomedical researchers can leverage the scale provided by high-performance computer and cloud storage facilities and national high-speed research networks while preserving privacy and meeting regulatory requirements.

Peisert, Sean↗

Evaluating integration and performance of containerized climate applications on a Hewlett Packard Enterprise Cray system

Containers have taken over large swaths of cloud computing as the most convenient way of packaging and deploying applications. The features that containers offer for packaging and deploying applications translate to high performance computing (HPC) as well. At The National Oceanic and Atmospheric Administration, containers provide an easy way to build and distribute complex HPC applications, allowing faster collaboration, portability, and experiment computer environment reproducibility amongst the scientific community. The challenge arises when applications rely on message passing interface (MPI). This necessitates investigation into how to properly run these applications with their own unique requirements and produce performance on par with native runs. We investigate the MPI performance for benchmarks and containerized climate models for various containers covering selection of compiler and MPI library combinations from the Cray provided programming environments on the Cray XC supercomputer GAEA. Performance from the benchmarks and the climate models shows that for the most part containerized applications perform on par with the natively built applications when the system optimized Cray MPICH libraries are bound into the container, and the hybrid model containers have poor performance in comparison. We also describe several challenges and our solutions in running these containers, particularly challenges with heterogeneous jobs for the containerized model runs.

Abraham, Subil↗

Quality and Performance of LED Flashlights in Kenya: Common End User Preferences and Complaints

Flashlights that use LED technology have quickly emerged as the dominant source of portable lighting in Kenya. While flashlights do not normally provide a substitute for kerosene and other highly inefficient fuels, they are an important early manifestation of LED lighting in the developing world that can serve as a platform - or deterrent - to the diffusion of the technology into the broader off-grid lighting market. The lead acid batteries embedded in flashlights also represent an important source of hazardous waste, and flashlight durability is thus an important determinant of the rate of waste disposal. Low-cost LED flashlights with prices from $1 to $4 are now widely available in shops and markets throughout Kenya. The increased penetration of LED technology in the flashlight market is significant, as over half of all Kenyan households report owning a flashlight (Kamfor, 2002). While this shift from conventional incandescent technology to modern LEDs may appear to be a promising development, end users that our research team interviewed expressed a number of complaints about the quality and performance of these new flashlights. This raises concerns about the interests of low-income flashlight users, and it may also indicate the onset of a broader market spoiling effect for off-grid lighting products based on LED technology (Mills and Jacobson, 2008; Lighting Africa, 2007). The quality of low-cost LED flashlights can contribute to market spoiling because these products appear to represent the first contact that most Kenyans have with LED technology. In this report, our team uses interviews with 46 end users of flashlights to collect information about their experiences, perceptions, and preferences. We focus especially on highlighting common complaints from respondents about the flashlights that they have used, as well as on noting the features that they indicated were important when evaluating the quality of a flashlight. In previous laboratory tests, researchers from our team found a wide range of quality and performance among battery powered LED lights (Granderson, et al. 2008).

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Spontaneous sodium ion storage behaviors of reduced graphene oxide anodes exceeding 100% Coulombic efficiency by modulated ion solvation

Rechargeable batteries are essential energy storage devices that power portable devices and electrical vehicles throughout the world. In general, it is thought that the electrochemical performance of rechargeable batteries is mostly determined by the electrodes within them and that the electrolyte plays a relatively passive role. However, ion transport and storage can be greatly influenced by the electrolyte solution structure, specifically, ion solvation within the bulk and ion desolvation across the electrode/electrolyte interfaces. Herein, we studied the role of the electrolyte as an active component of electrochemical energy storage devices. We found that with an appropriate electrolyte formulation, ion storage in disordered carbonaceous anode materials can occur spontaneously without externally supplied electrical energy. Reduced graphene oxide (RGO) in an ether-based electrolyte demonstrates ‘spontaneous' ion storage behaviors of adsorbing and inserting the solvated ions utilizing facilitated permeability and wettability of RGO, which results in Coulombic efficiency of ~145% due to additional charging capacity of ~180 mAh g -1 during electrochemical processes. The unexpected spontaneous ion storage behavior was extensively investigated using a combination of electrochemical analyses and diagnostics, advanced characterizations, and computational simulation. In conclusion, we believe the spontaneous ion storage behavior offers a new way to further improve the energy efficiency of practical rechargeable batteries.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Modeling subsurface reactive flows using leadership-class computing

We describe our experiences running PFLOTRAN–a code for simulation of coupled hydro-thermal-chemical processes in variably saturated, non-isothermal, porous media– on leadership-class supercomputers, including initial experiences running on the petaflop incarnation of Jaguar, the Cray XT5 at the National Center for Computational Sciences at Oak Ridge National Laboratory. PFLOTRAN utilizes fully implicit time-stepping and is built on top of the Portable, Extensible Toolkit for Scientific Computation (PETSc). We discuss some of the hurdles to "at scale" performance with PFLOTRAN and the progress we have made in overcoming them on leadership-class computer architectures.

Computer Science↗

Design considerations for future planetary space suits

Manned extravehicular activity (EVA) operations will be major mission elements of planned future U.S. space operations. Whether designed for orbital operations or planetary surface exploration, the EVA system must be safe and reliable, and must provide a high degree of performance capabilities. An extravehicular mobility unit (EMU) consisting of a space suit, EVA gloves, and a portable life support system (PLSS) is central to the EVA system. A rugged, highly reliable, mobile, reusable, and easily maintained EVA suit and compact PLSS must meet the specific requirements of the intended mission. Additional requirements imposed by exposure to surface and gravitational environments are the need for lightweight, high-strength materials for fabricating EMUs to prevent astronaut fatigue and the need for dust protection measures and removal techniques to prevent contamination. Also, operational pressure for habitats should be chosen so as to maintain zero-prebreathe conditions while allowing for lower EMU operating pressure.

Kosmo, Joseph J.↗

Development and Applications of a Modular Parallel Process for Large Scale Fluid/Structures Problems

A modular process that can efficiently solve large scale multidisciplinary problems using massively parallel super computers is presented. The process integrates disciplines with diverse physical characteristics by retaining the efficiency of individual disciplines. Computational domain independence of individual disciplines is maintained using a meta programming approach. The process integrates disciplines without affecting the combined performance. Results are demonstrated for large scale aerospace problems on several supercomputers. The super scalability and portability of the approach is demonstrated on several parallel computers.

Guruswamy, Guru P.↗

Development and Applications of a Modular Parallel Process for Large Scale Fluid/Structures Problems

A modular process that can efficiently solve large scale multidisciplinary problems using massively parallel supercomputers is presented. The process integrates disciplines with diverse physical characteristics by retaining the efficiency of individual disciplines. Computational domain independence of individual disciplines is maintained using a meta programming approach. The process integrates disciplines without affecting the combined performance. Results are demonstrated for large scale aerospace problems on several supercomputers. The super scalability and portability of the approach is demonstrated on several parallel computers.

Guruswamy, Guru P.↗

A Mathematical Model to Predict and Maintain the Neutral Buoyancy of Suited Astronauts

A previous study reported that inadequate weigh outs of suited subjects contribute to fatigue and the risk of injury during training in the Neutral Buoyancy Laboratory (NBL). Another study suggested that shoulder injuries observed in suited subjects who train in the NBL may be attributed to excessive righting moments caused by a non-optimal weigh out. The purpose of this study was to develop a mathematical model to predict and maintain the neutral buoyancy of suited subjects during training operations at the NBL. Due to time constraints, one certified NBL support diver served as a subject (height: 66.54 in; weight: 182 lbs) for this study and only one complete test was conducted. The study was divided into two runs for which the first run required the NBL divers to perform a weigh out similar to a suited astronaut on a scuba diver wearing a mock Portable Life Support System and a Displays and Control Module. For the second run, the same subject and equipment were weighed out according to the mathematical model. The objective of each run was to achieve a neutrally buoyant subject floating 450 to the pool floor. Motion data was collected using two underwater cameras and analyzed using Dartfish video analysis software while force and moment data were recorded using an AMTI force plate. The results from the NBL divers visual run indicate that the subject was floating at an angle of 29.50 while the resultant force and moment data were 1.139 lb and 1.125 ft-lb respectively. The mathematical model s weigh out resulted in the subject floating at an angle of 37.40 and a resultant force of 0.765 lb and resultant moment of 1.248 ft-lb. The mathematical model was better able to orient the subject and reduce resultant moment and force as compared to the NBL divers.

Clowers, Kurt↗

Silicon validation of evolution-designed circuits

This paper is the first to report on a silicon implementation of circutis evolved in simulation. Several circuits were evolved an fabricated in 0.5 micron CMOS process; this paper focuses on results of logical gates evolved at transistor level. It discusses the steps taken in order to increase the chances of robust and portable designs, summarizes the results of characterization tests based on chip measurements, and comments on the performance comparing to simulations.

evolutionary algorithms↗