Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Parallel and high performance computing”

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 829 records · Page 46
All sourcesNASA NTRSDOE OSTI
NASA NTRS1990

A real time neural net estimator of fatigue life

A neural net architecture is proposed to estimate, in real-time, the fatigue life of mechanical components, as part of the Intelligent Control System for Reusable Rocket Engines. Arbitrary component loading values were used as input to train a two hidden-layer feedforward neural net to estimate component fatigue damage. The ability of the net to learn, based on a local strain approach, the mapping between load sequence and fatigue damage has been demonstrated for a uniaxial specimen. Because of its demonstrated performance, the neural computation may be extended to complex cases where the loads are biaxial or triaxial, and the geometry of the component is complex (e.g., turbopump blades). The generality of the approach is such that load/damage mappings can be directly extracted from experimental data without requiring any knowledge of the stress/strain profile of the component. In addition, the parallel network architecture allows real-time life calculations even for high frequency vibrations. Owing to its distributed nature, the neural implementation will be robust and reliable, enabling its use in hostile environments such as rocket engines. This neural net estimator of fatigue life is seen as the enabling technology to achieve component life prognosis, and therefore would be an important part of life extending control for reusable rocket engines.

Troudet, T.↗
NASA NTRS1997

Opus: A Coordination Language for Multidisciplinary Applications

Data parallel languages, such as High Performance fortran, can be successfully applied to a wide range of numerical applications. However, many advanced scientific and engineering applications are multidisciplinary and heterogeneous in nature, and thus do not fit well into the data parallel paradigm. In this paper we present Opus, a language designed to fill this gap. The central concept of Opus is a mechanism called ShareD Abstractions (SDA). An SDA can be used as a computation server, i.e., a locus of computational activity, or as a data repository for sharing data between asynchronous tasks. SDAs can be internally data parallel, providing support for the integration of data and task parallelism as well as nested task parallelism. They can thus be used to express multidisciplinary applications in a natural and efficient way. In this paper we describe the features of the language through a series of examples and give an overview of the runtime support required to implement these concepts in parallel and distributed environments.

Chapman, Barbara↗
DOE OSTI2021

Performance of Julia for High Energy Physics Analyses

We argue that the Julia programming language is a compelling alternative to currently more common implementations in Python and C++ for common data analysis workflows in high energy physics. We compare the speed of implementations of different workflows in Julia with those in Python and C++. Furthermore, our studies show that the Julia implementations are competitive for tasks that are dominated by computational load rather than data access. For work that is dominated by data access, we demonstrate an application with concurrent file reading and parallel data processing.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Hvac: Removing I/O Bottleneck for Large-Scale Deep Learning Applications

Scientific communities are increasingly adopting deep learning (DL) models in their applications to accelerate scientific discovery processes. However, with rapid growth in the computing capabilities of HPC supercomputers, large-scale DL applications have to spend a significant portion of training time performing I/O to a parallel storage system. Previous research works have investigated optimization techniques such as prefetching and caching. Unfortunately, there exist non-trivial challenges to adopting the existing solutions on HPC supercomputers for large-scale DL training applications, which include non-performance and/or failures at extreme scale, lack of portability and generality in design, complex deployment methodology, and being limited to a specific application or dataset. To address these challenges, we propose High-Velocity AI Cache (HVAC), a distributed read-cache layer that targets and fully exploits the node-local storage or near node-local storage technology. HVAC seamlessly accelerates read I/O by aggregating node-local or near node-local storage, avoiding metadata lookups and file locking while preserving portability in the application code. We deploy and evaluate HVAC on 1,024 nodes (with over 6000 NVIDIA V100 GPUS) of the Summit supercomputer. In particular, we evaluate the scalability, efficiency, accuracy, and load distribution of HVAC compared to GPFS and XFS-on-NVMe. With four different DL applications, we observe an average 25 % performance improvement atop GPFS and 9% drop against XFS-on-NVMe, which scale linearly and are considered the performance upper bound. We envision HVAC as an important caching library for upcoming HPC supercomputers such as Frontier.

Khan, Awais↗
DOE OSTI2026

A scalable framework for efficient coupling of thermal and microstructural simulations in additive manufacturing

Predicting microstructure evolution in metal additive manufacturing (AM) is important for process optimization, but spatiotemporal scale disparities between thermal transport and microstructure evolution create significant challenges for efficient data transfer between simulation codes. To address this, we present Stork, a scalable framework for coupling thermal and microstructural simulations. Stork uses a sparse data representation to identify and store active solidification sub-volumes, enabling highly parallel quad-linear interpolation from coarse thermal grids to fine microstructure grids without large intermediate storage. We demonstrate the framework by coupling the semi-analytic heat transfer code 3DThesis with the time-parallel cellular automata code Toucan. This approach achieves over two orders of magnitude reduction in data generation time and file size compared to prior workflows. Numerical studies show that quad-linear interpolation preserves grain morphology and crystallographic texture in laser powder bed fusion (LPBF) simulations for coarsening ratios up to 16. Overall, Stork provides a scalable pathway for high-throughput, component-scale AM simulations on modern high-performance computing systems.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Optimizing Error-Bounded Lossy Compression for Scientific Data on GPUs

Error-bounded lossy compression is a critical technique for significantly reducing scientific data volumes. With ever-emerging heterogeneous high-performance computing (HPC) architecture, GPU-accelerated error-bounded compressors (such as CUSZ and cuZFP) have been developed. However, they suffer from either low performance or low compression ratios. To this end, we propose CUSZ+ to target both high compression ratios and throughputs. We identify that data sparsity and data smoothness are key factors for high compression throughputs. Our key contributions in this work are fourfold: (1) We propose an efficient compression workflow to adaptively perform run-length encoding and/or variable-length encoding. (2) We derive Lorenzo reconstruction in decompression as multidimensional partial-sum computation and propose a fine-grained Lorenzo reconstruction algorithm for GPU architectures. (3) We carefully optimize each of CUSZ kernels by leveraging state-of-the-art CUDA parallel primitives. (4) We evaluate CUSZ+ using seven real-world HPC application datasets on V100 and A100 GPUs. Experiments show CUSZ+ improves the compression throughputs and ratios by up to 18.4x and 5.3x, respectively, over CUSZ on the tested datasets.

Tian, Jiannan↗
DOE OSTI2021

Accelerating Binarized Neural Networks via Bit-Tensor-Cores in Turing GPUs

Despite foreseeing tremendous speedups over conventional deep neural networks, the performance advantage of binarized neural networks (BNNs) has merely been showcased on general-purpose processors such as CPUs and GPUs. In fact, due to being unable to leverage bit-level-parallelism with a word-based architecture, GPUs have been criticized for extremely low utilization (1%) when executing BNNs. Consequently, the latest tensorcores in NVIDIA Turing GPUs start to experimentally support bit computation. In this work, we look into this brand new bit computation capability and characterize its unique features. We show that the stride of memory access can significantly affect performance delivery and a data-format co-design is highly desired to support the tensorcores for achieving superior performance than existing software solutions without tensorcores. We realize the tensorcore-accelerated BNN design, particularly the major functions for fully-connect and convolution layers — bit matrix multiplication and bit convolution. Evaluations on two NVIDIA Turing GPUs show that, with ResNet-18, our BTC-BNN design can process ImageNet at a rate of 5.6K images per second, 77% faster than state-of-the-art. Our BNN approach is released on https://github.com/pnnl/TCBNN.

Li, Ang↗
NASA NTRS1990

A real time neural net estimator of fatigue life

A neural network architecture is proposed to estimate, in real-time, the fatigue life of mechanical components, as part of the intelligent Control System for Reusable Rocket Engines. Arbitrary component loading values were used as input to train a two hidden-layer feedforward neural net to estimate component fatigue damage. The ability of the net to learn, based on a local strain approach, the mapping between load sequence and fatigue damage has been demonstrated for a uniaxial specimen. Because of its demonstrated performance, the neural computation may be extended to complex cases where the loads are biaxial or triaxial, and the geometry of the component is complex (e.g., turbopumps blades). The generality of the approach is such that load/damage mappings can be directly extracted from experimental data without requiring any knowledge of the stress/strain profile of the component. In addition, the parallel network architecture allows real-time life calculations even for high-frequency vibrations. Owing to its distributed nature, the neural implementation will be robust and reliable, enabling its use in hostile environments such as rocket engines.

Troudet, T.↗
NASA NTRS1992

Network issues for large mass storage requirements

File Servers and Supercomputing environments need high performance networks to balance the I/O requirements seen in today's demanding computing scenarios. UltraNet is one solution which permits both high aggregate transfer rates and high task-to-task transfer rates as demonstrated in actual tests. UltraNet provides this capability as both a Server-to-Server and Server-to-Client access network giving the supercomputing center the following advantages highest performance Transport Level connections (to 40 MBytes/sec effective rates); matches the throughput of the emerging high performance disk technologies, such as RAID, parallel head transfer devices and software striping; supports standard network and file system applications using SOCKET's based application program interface such as FTP, rcp, rdump, etc.; supports access to the Network File System (NFS) and LARGE aggregate bandwidth for large NFS usage; provides access to a distributed, hierarchical data server capability using DISCOS UniTree product; supports file server solutions available from multiple vendors, including Cray, Convex, Alliant, FPS, IBM, and others.

Perdue, James↗
NASA NTRS1993

Special purpose parallel computer architecture for real-time control and simulation in robotic applications

This is a real-time robotic controller and simulator which is a MIMD-SIMD parallel architecture for interfacing with an external host computer and providing a high degree of parallelism in computations for robotic control and simulation. It includes a host processor for receiving instructions from the external host computer and for transmitting answers to the external host computer. There are a plurality of SIMD microprocessors, each SIMD processor being a SIMD parallel processor capable of exploiting fine grain parallelism and further being able to operate asynchronously to form a MIMD architecture. Each SIMD processor comprises a SIMD architecture capable of performing two matrix-vector operations in parallel while fully exploiting parallelism in each operation. There is a system bus connecting the host processor to the plurality of SIMD microprocessors and a common clock providing a continuous sequence of clock pulses. There is also a ring structure interconnecting the plurality of SIMD microprocessors and connected to the clock for providing the clock pulses to the SIMD microprocessors and for providing a path for the flow of data and instructions between the SIMD microprocessors. The host processor includes logic for controlling the RRCS by interpreting instructions sent by the external host computer, decomposing the instructions into a series of computations to be performed by the SIMD microprocessors, using the system bus to distribute associated data among the SIMD microprocessors, and initiating activity of the SIMD microprocessors to perform the computations on the data by procedure call.

Fijany, Amir↗
NASA NTRS1995

Algorithms for parallel flow solvers on message passing architectures

The purpose of this project has been to identify and test suitable technologies for implementation of fluid flow solvers -- possibly coupled with structures and heat equation solvers -- on MIMD parallel computers. In the course of this investigation much attention has been paid to efficient domain decomposition strategies for ADI-type algorithms. Multi-partitioning derives its efficiency from the assignment of several blocks of grid points to each processor in the parallel computer. A coarse-grain parallelism is obtained, and a near-perfect load balance results. In uni-partitioning every processor receives responsibility for exactly one block of grid points instead of several. This necessitates fine-grain pipelined program execution in order to obtain a reasonable load balance. Although fine-grain parallelism is less desirable on many systems, especially high-latency networks of workstations, uni-partition methods are still in wide use in production codes for flow problems. Consequently, it remains important to achieve good efficiency with this technique that has essentially been superseded by multi-partitioning for parallel ADI-type algorithms. Another reason for the concentration on improving the performance of pipeline methods is their applicability in other types of flow solver kernels with stronger implied data dependence. Analytical expressions can be derived for the size of the dynamic load imbalance incurred in traditional pipelines. From these it can be determined what is the optimal first-processor retardation that leads to the shortest total completion time for the pipeline process. Theoretical predictions of pipeline performance with and without optimization match experimental observations on the iPSC/860 very well. Analysis of pipeline performance also highlights the effect of uncareful grid partitioning in flow solvers that employ pipeline algorithms. If grid blocks at boundaries are not at least as large in the wall-normal direction as those immediately adjacent to them, then the first processor in the pipeline will receive a computational load that is less than that of subsequent processors, magnifying the pipeline slowdown effect. Extra compensation is needed for grid boundary effects, even if all grid blocks are equally sized.

Vanderwijngaart, Rob F.↗
DOE OSTI2020

Gradient-Based Multi-Area Distribution System State Estimation

The increasing distributed and renewable energy resources and controllable devices in distribution systems make fast distribution system state estimation (DSSE) crucial in system monitoring and control. We consider a large multi-phase distribution system and formulate DSSE as a weighted least squares (WLS) problem. We divide the large distribution system into smaller areas of subtree structure, and by jointly exploring the linearized power flow model and the network topology, we propose a gradient-based multi-area algorithm to exactly and efficiently solve the WLS problem. The proposed algorithm enables distributed and parallel computation of the state estimation problem without compromising any performance. Numerical results on a 4,521-node test feeder show that the designed algorithm features fast convergence and accurate estimation results. Comparison with traditional Gauss-Newton method shows that the proposed method has much better performance in distribution systems with a limited amount of reliable measurement. The real-time implementation of the algorithm tracks time-varying system states with high accuracy.

41 EE - Solar Energy Technologies Office (EE-4S)↗
NASA NTRS1993

Real-time dynamics simulation of the Cassini spacecraft using DARTS. Part 1: Functional capabilities and the spatial algebra algorithm

This paper describes the Dynamics Algorithms for Real-Time Simulation (DARTS) real-time hardware-in-the-loop dynamics simulator for the National Aeronautics and Space Administration's Cassini spacecraft. The spacecraft model consists of a central flexible body with a number of articulated rigid-body appendages. The demanding performance requirements from the spacecraft control system require the use of a high fidelity simulator for control system design and testing. The DARTS algorithm provides a new algorithmic and hardware approach to the solution of this hardware-in-the-loop simulation problem. It is based upon the efficient spatial algebra dynamics for flexible multibody systems. A parallel and vectorized version of this algorithm is implemented on a low-cost, multiprocessor computer to meet the simulation timing requirements.

Jain, A.↗
NASA NTRS2001

Comparison of Implicit Collocation Methods for the Heat Equation

We combine a high-order compact finite difference scheme to approximate spatial derivatives arid collocation techniques for the time component to numerically solve the two dimensional heat equation. We use two approaches to implement the collocation methods. The first one is based on an explicit computation of the coefficients of polynomials and the second one relies on differential quadrature. We compare them by studying their merits and analyzing their numerical performance. All our computations, based on parallel algorithms, are carried out on the CRAY SV1.

Kouatchou, Jules↗
DOE OSTI2024

Accelerating Multivariate Functional Approximation Computation with Domain Decomposition Techniques⋆

Modeling large datasets through Multivariate Functional Approximations (MFA) provide an elegant way to handle many visualization and scientific analysis workflows. The process necessitates scalable data partitioning methods to compute MFA representations efficiently without compromising the accuracy or continuity of the reconstructed solution. We propose a domain -decomposed method for computing the MFA with B -spline bases, which reduces the total work per task and uses a restricted Additive Schwarz (RAS) method to converge the control point data degrees -of -freedom along subdomain boundaries. We provide an in-depth analysis of the parallel approach with domain decomposition solvers, aiming to minimize local subdomain error residuals and recover high -order continuity at subdomain interfaces with appropriate choices of knot overlaps. The communication cost, determined by the overlap regions in the RAS implementation, is optimized to recover the numerical error profile of the single subdomain case. Our proposed method stands in contrast to previous methods, which typically only recover either C 0 or at best C 1 continuity for arbitrary B -spline degree expansions, or those that require post -processing to blend discontinuities in the reconstructed data. We demonstrate the effectiveness of our approach using analytical and real -world datasets in 1D, 2D, and 3D through both strong and weak scaling studies. The performance results indicate that the overall cost of computing the approximation is directly proportional to the underlying nearest -neighbor communication implementation, and is only weakly dependent on the overlap region size that determines the size of the messages. This finding underscores the efficiency and scalability of our proposed method, making it a promising solution for handling large datasets in scientific workflows.

additive Schwarz solvers↗
NASA NTRS1992

High performance compression of science data

In the future, NASA expects to gather over a tera-byte per day of data requiring space for levels of archival storage. Data compression will be a key component in systems that store this data (e.g., optical disk and tape) as well as in communications systems (both between space and Earth and between scientific locations on Earth). We propose to develop algorithms that can be a basis for software and hardware systems that compress a wide variety of scientific data with different criteria for fidelity/bandwidth tradeoffs. The algorithmic approaches we consider are specially targeted for parallel computation where data rates of over 1 billion bits per second are achievable with current technology.

Storer, James A.↗
NASA NTRS1993

High performance compression of science data

In the future, NASA expects to gather over a tera-byte per day of data requiring space for levels of archival storage. Data compression will be a key component in systems that store this data (e.g., optical disk and tape) as well as in communications systems (both between space and Earth and between scientific locations on Earth). We propose to develop algorithms that can be a basis for software and hardware systems that compress a wide variety of scientific data with different criteria for fidelity/bandwidth tradeoffs. The algorithmic approaches we consider are specially targeted for parallel computation where data rates of over 1 billion bits per second are achievable with current technology.

Storer, James A.↗
NASA NTRS2015

Vibroacoustic Tailoring of a Rod-Stiffened Composite Fuselage Panel with Multidisciplinary Considerations

An efficient multi-objective design tailoring procedure seeking to improve the vibroacoustic performance of a fuselage panel while maintaining or reducing weight is presented. The structure considered is the pultruded rod stitched efficient unitized structure, a highly integrated composite structure concept designed for a noncylindrical, next-generation flight vehicle fuselage. Modifications to a baseline design are evaluated within a six-parameter design space including spacing, flange width, and web height for both frame and stringer substructure components. The change in sound power radiation attributed to a design change is predicted using finite-element models sized and meshed for analyses in the 500 Hz, 1 kHz, and 2 kHz octave bands. Three design studies are carried out in parallel while considering a diffuse acoustic field excitation and two types of turbulent boundary-layer excitation. Kriging surrogate models are used to reduce the computational costs of resolving the vibroacoustic and weight objective Pareto fronts. The resulting Pareto optimal designs are then evaluated under a static pressurization ultimate load to assess structural strength and stability. Results suggest that choosing alternative configurations within the considered design space can reduce weight and improve vibroacoustic performance without compromising strength and stability of the structure under the static load condition considered, but the tradeoffs are significantly influenced by the spatial characteristics of the assumed excitation field.

Allen, Albert R.↗