Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Irregular applications”

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 199 records · Page 11

Concurrency-based approaches to parallel programming

The inevitable transition to parallel programming can be facilitated by appropriate tools, including languages and libraries. After describing the needs of applications developers, this paper presents three specific approaches aimed at development of efficient and reusable parallel software for irregular and dynamic-structured problems. A salient feature of all three approaches in their exploitation of concurrency within a processor. Benefits of individual approaches such as these can be leveraged by an interoperability environment which permits modules written using different approaches to co-exist in single applications.

Kale, L.V.↗

Aperture‐Synthesis Radar Imaging With Compressive Sensing for Ionospheric Research

Abstract Inverse methods involving compressive sensing are tested in the application of two‐dimensional aperture‐synthesis imaging of radar backscatter from field‐aligned plasma density irregularities in the ionosphere. We consider basis pursuit denoising, implemented with the fast iterative shrinkage thresholding algorithm, and orthogonal matching pursuit (OMP) with a wavelet basis in the evaluation. These methods are compared with two more conventional optimization methods rooted in entropy maximization (MaxENT) and adaptive beamforming (linearly constrained minimum variance or often “Capon's Method.”) Synthetic data corresponding to an extended ionospheric radar target are considered. We find that MaxENT outperforms the other methods in terms of its ability to recover imagery of an extended target with broad dynamic range. Fast iterative shrinkage thresholding algorithm performs reasonably well but does not reproduce the full dynamic range of the target. It is also the most computationally expensive of the methods tested. OMP is very fast computationally but prone to a high degree of clutter in this application. We also point out that the formulation of MaxENT used here is very similar to OMP in some respects, the difference being that the former reconstructs the logarithm of the image rather than the image itself from basis vectors extracted from the observation matrix. MaxENT could in that regard be considered a form of compressive sensing.

Hysell, D. L.↗

A Fault-oblivious Extreme-scale Execution Environment

Exascale computing systems will provide a thousand-fold increase in parallelism and a proportional increase in failure rate relative to today's machines. Systems software for exascale machines must provide the infrastructure to support existing applications while simultaneously enabling efficient execution of new programming models that naturally express dynamic, adaptive, irregular computation; coupled simulations; and massive data analysis in a highly unreliable hardware environment with billions of threads of execution. We propose a new approach to the data and work distribution model provided by system software based on the unifying formalism of an abstract file system. The proposed hierarchical data model provides simple, familiar visibility and access to data structures through the file system hierarchy, while providing fault tolerance through selective redundancy. The hierarchical task model features work queues whose form and organization are represented as file system objects. Data and work are both first class entities. By exposing the relationships between data and work to the runtime system, information is available to optimize execution time and provide fault tolerance. The data distribution scheme provides replication (where desirable and possible) for fault tolerance and efficiency, and it is hierarchical to make it possible to take advantage of locality. The user, tools, and applications, including legacy applications, can interface with the data, work queues, and one another through the abstract file model. This runtime environment will provide multiple interfaces to support traditional Message Passing Interface applications, languages developed under DARPA's High Productivity Computing Systems program, as well as other, experimental programming models. We will validate our runtime system with pilot codes on existing platforms and will use simulation to validate for exascale-class platforms. In this final report, we summarize research results from the work done at the Ohio State University towards the larger goals of the project listed above.

97 MATHEMATICS AND COMPUTING↗

Considerations on the Use of Custom Accelerators for Big Data Analytics

Accelerators, including Graphic Processing Units (GPUs) for gen- eral purpose computation, many-core designs with wide vector units (e.g., Intel Phi), have become a common component of many high performance clusters. The appearance of more stable and reliable tools tools that can automatically convert code written in high-level specifications with annotations (such as C or C++) to hardware de- scription languages (High-Level Synthesis - HLS), is also setting the stage for a broader use of reconfigurable devices (e.g., Field Pro- grammable Gate Arrays - FPGAs) in high performance system for the implementation of custom accelerators, helped by the fact that new processors include advanced cache-coherent interconnects for these components. In this chapter, we briefly survey the status of the use of accelerators in high performance systems targeted at big data analytics applications. We argue that, although the progress in the use of accelerators for this class of applications has been sig- nificant, differently from scientific simulations there still are gaps to close. This is particularly true for the ”irregular” behaviors exhibited by no-SQL, graph databases. We focus our attention on the limits of HLS tools for data analytics and graph methods, and discuss a new architectural template that better fits the requirement of this class of applications. We validate the new architectural templates by mod- ifying the Graph Engine for Multithreaded System (GEMS) frame- work to support accelerators generated with such a methodology, and testing with queries coming from the Lehigh University Benchmark (LUBM). The architectural template enables better supporting the task and memory level parallelism present in graph methods by sup- porting a new control model and a enhanced memory interface. We show that out solution allows generating parallel accelerators, pro- viding speed ups with respect to conventional HLS flows. We finally draw conclusions and present a perspective on the use of reconfig- urable devices and Design Automation tools for data analytics.

Data Analytics↗

Electromagnetic Scattering From a Polygonal Thin Metallic Plate Using Quadrilateral Meshing

The problem of electromagnetic (EM) scattering from irregularly shaped, thin, metallic flat plates in free space is solved using the electric field integral equation (EFIE) approach in conjunction with the method of moments (MoM) with quadrilateral meshing. An irregularly shaped thin plate is discretized into quadrilateral patches and the unknown electric surface current over the plate is expressed in terms of proper basis functions over these patches. The basis functions for the electric surface current density that satisfy the proper boundary conditions on these quadrilateral patches are derived. The unknown surface current density on these quadrilateral patches is determined by setting up and solving the electric field integral equation by the application of the MoM. From the knowledge of the surface current density, the EM scattering from various irregularly shaped plates is determined and compared with the earlier published results. The novelty in the present approach is the use of quadrilateral patches instead of well known and often used triangular patches. The numerical results obtained using the quadrilateral patches compare favorably with measured results.

Deshpande, Manohar D.↗

Dynamic Load Balancing on Single- and Multi-GPU Systems

The computational power provided by many-core graphics processing units (GPUs) has been exploited in many applications. The programming techniques supported and employed on these GPUs are not sufficient to address problems exhibiting irregular, and unbalanced workload. The problem is exacerbated when trying to effectively exploit multiple GPUs, which are commonly available in many modern systems. In this paper, we propose a task-based dynamic load-balancing solution for single- and multi-GPU systems. The solution allows load balancing at a finer granularity than what is supported in existing APIs such as NVIDIA’s CUDA. We evaluate our approach using both micro-benchmarks and a molecular dynamics application that exhibits significant load imbalance. Experimental results with a single-GPU configuration show that our fine-grained task solution can utilize the hardware more efficiently than the CUDA scheduler for unbalanced workload. On multi-GPU systems, our solution achieves near-linear speedup, load balance, and significant performance improvement over techniques based on standard CUDA APIs.

97 MATHEMATICS AND COMPUTING↗

The relationship between strength of turbulence and backscattering radar power at HF and VHF

The formulae relating turbulence and other atmospheric parameters to backscattered power for radar observations are reviewed. Emphasis is on the case of scatter from turbulent irregularities which have scales corresponding to the range of isotropic, inertial range turbulence. The applicability of this assumption is discussed. A formula is introduced for the mesosphere which relates ionospheric electron densities to backscattered power.

Hocking, W. K.↗

Modelling of microwave emission and scattering from snow and soil

In the past a snow layer has been modeled as a homogeneous layer embedded with sparsely populated Rayleigh scatterers above an irregular ground surface. The effect of the ground surface can be ignored if the layer is sufficiently lossy due to wetness in the snow. The top surface of the snow layer may be treated as plane or irregular depending upon its actual shape and its wetness condition. For a dry snow condition where the electromagnetic wave can penetrate easily one can ignore the air-snow interface. As a result a variety of emission and scattering models exist. An improvement to the existing scattering or emission model would consist of an irregular layer with densely populated correlated scatterers. The development of this model and its application to scattering and emission from a snow layer are discussed. Also disucssed is a surface scattering model for a soil surface.

Fung, Adrian K.↗

Performance Analysis of Scientific Computing Workloads on Trusted Execution Environments

Scientific computing sometimes involves computation on sensitive data. Depending on the data and the execution environment, the HPC (high-performance computing) user or data provider may require confidentiality and/or integrity guarantees. To study the applicability of hardware-based trusted execution environments (TEEs) to enable secure scientific computing, we deeply analyze the performance impact of AMD SEV and Intel SGX for diverse HPC benchmarks including traditional scientific computing, machine learning, graph analytics, and emerging scientific computing workloads. We observe three main findings: 1) SEV requires careful memory placement on large scale NUMA machines (1x -3.4x slowdown without and 1x -1.15x slowdown with NUMA aware placement), 2) virtualization - a prerequisite for SEV - results in performance degradation for workloads with irregular memory accesses and large working sets (1x -4x slowdown compared to native execution for graph applications) and 3) SGX is inappropriate for HPC given its limited secure memory size and inflexible programming model (1.2x -126x slowdown over unsecure execution). Finally, we discuss forthcoming new TEE designs and their potential impact on scientific computing.

97 MATHEMATICS AND COMPUTING↗

Low subsonic aerodynamic characteristics of five irregular planform wings with systematically varying wing fillet geometry tested in the NASA/Ames 12 foot pressure tunnel (LA65)

An experimental and analytical aerodynamic program to develop predesign guides for irregular planform wings (also referred to as cranked leading edge or double delta wings is reported; the benefits are linearization of subsonic lift curve slope to high angles of attack and avoidance of subsonic pitch instabilities at high lift by proper tailoring of the planform-fillet-wing combination while providing the desired hypersonic trim angle and stability. Because subsonic and hypersonic conditions were the two prime areas of concern in the initial application of this program to optimize shuttle orbiter landing and entry characteristics, the study was designated the Subsonic/Hypersonic Irregular Planforms Study (SHIPS).

Ball, J. W.↗

Stochastic modelling of intermittent fluctuations in the scrape-off layer: Correlations, distributions, level crossings, and moment estimation

A stochastic model is presented for intermittent fluctuations in the scrape-off layer of magnetically confined plasmas. The fluctuations in the plasma density are modeled by a super-position of uncorrelated pulses with fixed shape and duration, describing radial motion of blob-like structures. In the case of an exponential pulse shape and exponentially distributed pulse amplitudes, predictions are given for the lowest order moments, probability density function, auto-correlation function, level crossings, and average times for periods spent above and below a given threshold level. Also, the mean squared errors on estimators of sample mean and variance for realizations of the process by finite time series are obtained. These results are discussed in the context of single-point measurements of fluctuations in the scrape-off layer, broad density profiles, and implications for plasma–wall interactions due to the transient transport events in fusion grade plasmas. The results may also have wide applications for modelling fluctuations in other magnetized plasmas such as basic laboratory experiments and ionospheric irregularities.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Defect Spirograph: Dynamical Behavior of Defects in Spatially Patterned Active Nematics

We report that topological defects in active liquid crystals can be confined by introducing gradients of activity. Here, we examine the dynamical behavior of two defects confined by a sharp gradient of activity that separates an active circular region and a surrounding passive nematic material. Continuum simulations are used to explain how the interplay among energy injection into the system, hydrodynamic interactions, and frictional forces governs the dynamics of topologically required self-propelling + 1/2 defects. Our findings are rationalized in terms of a phase diagram for the dynamical response of defects in terms of activity and frictional damping strength. Different regions of the underlying phase diagram correspond to distinct dynamical modes, namely immobile defects, steady rotation of defects, bouncing defects, bouncing-cruising defects, dancing defects, and multiple defects with irregular dynamics. These dynamic states raise the prospect of generating synchronized defect arrays for microfluidic applications.

36 MATERIALS SCIENCE↗

Local plate/rod descriptors of 3D trabecular bone micro-CT images from medial axis topologic analysis

Purpose: Trabecular bone microarchitecture is made of a complex network of plate and rod structures evolving with age and disease. The purpose of this article is to propose a new 3D local analysis method for the quantitative assessment of parameters related to the geometry of trabecular bone microarchitecture. Methods: The method is based on the topologic classification of the medial axis of the 3D image into branches, rods, and plates. Thanks to the reversibility of the medial axis, the classification is next extended to the whole 3D image. Finally, the percentages of rods and plates as well as their mean thicknesses are calculated. The method was applied both to simulated test images and 3D micro-CT images of human trabecular bone. Results: The classification of simulated phantoms made of plates and rods shows that the maximum error in the quantitative percentages of plate and rods is less than 6% and smaller than with the structure model index (SMI). Micro-CT images of human femoral bone taken in osteoporosis and early or advanced osteoarthritis were analyzed. Despite the large physiological variability, the present method avoids the underestimation of rods observed with other local methods. The relative percentages of rods and plates were not significantly different between osteoarthritis and osteoporotic groups, whereas their absolute percentages were in relation to an increase of rod and plate thicknesses in advanced osteoarthritis with also higher relative and absolute number of nodes. Conclusions: The proposed method is model-independent, robust to surface irregularities, and enables geometrical characterization of not only skeletal structures but entire 3D images. Its application provided more accurate results than the standard SMI on simple simulated phantoms, but the discrepancy observed on the advanced osteoarthritis group raises questions that will require further investigations. The systematic use of such a local method in the characterization of trabecular bone samples could provide new insight in bone microarchitecture changes related to bone diseases or to those induced by drugs or therapy.

60 APPLIED LIFE SCIENCES↗

Performance Analysis, Modeling and Scaling of HPC Applications and Tools

E cient use of supercomputers at DOE centers is vital for maximizing system throughput, mini- mizing energy costs and enabling science breakthroughs faster. This requires complementary e orts along several directions to optimize the performance of scienti c simulation codes and the under- lying runtimes and software stacks. This in turn requires providing scalable performance analysis tools and modeling techniques that can provide feedback to physicists and computer scientists developing the simulation codes and runtimes respectively. The PAMS project is using time allocations on supercomputers at ALCF, NERSC and OLCF to further the goals described above by performing research along the following fronts: 1. Scaling Study of HPC applications; 2. Evaluation of Programming Models; 3. Hardening of Performance Tools; 4. Performance Modeling of Irregular Codes; and 5. Statistical Analysis of Historical Performance Data. We are a team of computer and computational scientists funded by both DOE/NNSA and DOE/ ASCR programs such as ECRP, XStack (Traleika Glacier, PIPER), ExaOSR (ARGO), SDMAV II (MONA) and PSAAP II (XPACC). This allocation will enable us to study big data issues when analyzing performance on leadership computing class systems and to assist the HPC community in making the most e ective use of these resources.

97 MATHEMATICS AND COMPUTING↗

Design considerations for application of laminar flow control systems to transport aircraft

The current status of the laminar-flow control LFC technology is summarized. Factors that have previously inhibited the application of LFC are first reviewed. Involved are the effects of atmospheric ice crystals, surface irregularities, acoustical environment, and off-design operating conditions. Aircraft design trends that are different from turbulent aircraft are discussed as are various design requirements unique to the LFC systems. Current design approaches for the principal LFC systems are reviewed. These include the system for protection of the leading-edge region from surface contamination and icing and the system for removal of a portion of the boundary-layer air. The latter includes consideration of both multiple spanwise suction slots and distributed perforations and required differences between the wing-box and leading-edge box regions.

Braslow, A. L.↗

Using Multithreading for the Automatic Load Balancing of 2D Adaptive Finite Element Meshes

In this paper, we present a multi-threaded approach for the automatic load balancing of adaptive finite element (FE) meshes. The platform of our choice is the EARTH multi-threaded system which offers sufficient capabilities to tackle this problem. We implement the question phase of FE applications on triangular meshes, and exploit the EARTH token mechanism to automatically balance the resulting irregular and highly nonuniform workload. We discuss the results of our experiments on EARTH-SP2, an implementation of EARTH on the IBM SP2, with different load balancing strategies that are built into the runtime system.

Heber, Gerd↗

Using Multi-threading for the Automatic Load Balancing of 2D Adaptive Finite Element Meshes

In this paper, we present a multi-threaded approach for the automatic load balancing of adaptive finite element (FE) meshes The platform of our choice is the EARTH multi-threaded system which offers sufficient capabilities to tackle this problem. We implement the adaption phase of FE applications oil triangular meshes and exploit the EARTH token mechanism to automatically balance the resulting irregular and highly nonuniform workload. We discuss the results of our experiments oil EARTH-SP2, on implementation of EARTH on the IBM SP2 with different load balancing strategies that are built into the runtime system.

Heber, Gerd↗

DS-GL: Advancing Graph Learning via Harnessing the Power of Nature within Dynamic Systems

With the rapid digitization of the world, an increasing number of real-world applications are turning to nonEuclidean data, modeled as graphs. Due to their intrinsic high complexity and irregularity, learning from graph data demands tremendous computational power. Recently, CMOS-compatible Ising machines, i.e., dynamic systems composed of CMOS components, have emerged as a new approach that harnesses the inherent power of natural annealing within dynamic systems to efficiently resolve binary optimization problems and have been adopted for traditional graph computation, such as max-cut. However, when performing complex Graph Learning (GL) tasks, Ising machines face significant hurdles: (i) they are inherently binary and thus ill-suited for real-valued problems; (ii) their expensive all-to-all coupling network that guarantees effective natural annealing poses daunting scalability concerns. To address these challenges, this paper proposes a nature-powered graph learning framework dubbed DS-GL, which is the first effort to transform the process of solving graph learning problems into the natural annealing process within a parameterized dynamic system embodied as a CMOS chip. To tackle the two major hurdles, DS-GL first augments the Ising machine architecture to modify the self-reaction term of its Hamiltonian function from linear to quadratic, effectively serving as an energy regulator. This adjustment maintains the system’s original physical interpretation while enabling it to process continuous, real-valued data. Second, to address the scaling issue, DS-GL further upgrades the real-valued dense Ising machine by decomposing it into a mesh-based multi-PE dynamic system that supports efficient distributed spatial-temporal co-annealing across different PEs through sparse interconnects. By exploiting the inherent sparsity and component structures in real-world graphs, DS-GL is able to map complex graph learning tasks onto the scalable dynamic system while maintaining high accuracy. Evaluations with three diverse GL applications across six real-world datasets, including traffic flow and COVID-19 prediction, show that DS-GL can deliver from 102× to 106× speedups and 500× energy reduction over Graph Neural Networks on GPUs, with 5% - 20% accuracy enhancement.

Song, Ruibing↗