Search NASA⌕ Search

SEARCH · Search NASA

Results for “Memory Optimization”

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 901 records · Page 50
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

DTLMod: A simulation framework for in situ workflow optimization

In situ processing workflows have become essential for coping with the explosion in data volume and velocity in large-scale scientific computing, providing domain scientists with early insights at runtime. Multiple frameworks implement this paradigm through a data transport layer (DTL), offering different data access modes and deployment schemes, but researchers currently lack the appropriate tools to assess design and deployment options before committing to costly real experiments. We introduce DTLMod, an open-source simulated DTL that enables performance evaluation of in situ workflow configurations at scale. Built on SimGrid, it links into any SimGrid-based simulator and is available in C++ and Python. We evaluate DTLMod along four axes: scalability (tens of thousands of simulated processes across interconnected clusters in seconds, with linear memory scaling), versatility (three implementation variants trading fidelity for speed), accuracy (simulated times faithfully reflecting real behavior), and practical utility (two use cases demonstrating evidence-based workflow design decisions).

Suter, Fred [ORNL] (ORCID:0000000319021955)↗
DOE OSTI2018

DeStager: feature guided in-situ data management in distributed deep memory hierarchies

In-situ analytics have been increasingly adopted by leadership scientific applications to gain fast insights into massive output data of simulations. With the current practice, systems buffer the output data in DRAM for analytics processing, constraining it to DRAM capacity un-used by the simulation. The rapid growth of data size requires alternative approaches to accommodating data-rich analytics, such as using solid-state disks to increase effective memory capacity. For this purpose, this paper explores software solutions for exploring the deep memory hierarchies expected on future high-end machines. Leveraging the fact that many analytics are sensitive to data features (regions-of-interest) hidden in the data being processed, the approach incorporates the knowledge of the data features into in-situ data management. It uses adaptive index creation/refinement to reduce the overhead of index management. In addition, it uses data features to predict data skew and improve load balance through controlling data distribution and placement on distributed staging servers. The experimental results show that such feature-guided optimizations achieve substantial improvements over state-of-the-art approaches for managing output data in-situ.

Computer Science↗
DOE OSTI2018

Evaluating and Optimizing OpenCL Base64 Data Unpacking Kernel with FPGA

Development of applications using OpenCL targeting FPGAs is an emerging approach on heterogeneous computing systems. This paper uses the data unpacking algorithm in Base64 encoding as a case study to present programming and optimization techniques, and experimental results of the OpenCL-based implementations on an FPGA. We explain the algorithm and evaluate the performance of the kernel implementations with Intel's FPGA OpenCL SDK. The experimental results show kernel vectorization and duplication are two optimization techniques that can improve the kernel performance. The performance of kernel duplication is also closely related to the local work size. Our experiment shows 16-lane vectorization increases the bandwidth by a factor of 2 to 10 for large input data sizes. Moreover, the performance of kernel duplication using 16 compute units is 40% to 1.5% less than that of kernel vectorization depending on the input size. Tuning the local work size can improve the kernel performance by a factor of 3 to 23. For this kernel, using local memory is not an effective technique to improve the kernel performance because input data is not reused. A combination of vectorization and duplication achieves the highest performance of 12.3 GiB/s. Compared to an Intel Xeon E5 CPU and an Nvidia Tesla K80 GPU, the performance of the kernel on the Arria 10 FPGA is 6.7X faster than the CPU and 3X slower than the GPU. The performance per watt on the FPGA is 20.5X higher than the CPU and 1.19X lower than the GPU.

Base64 Encoding↗
DOE OSTI2022

Ultrafast Error-bounded Lossy Compression for Scientific Datasets

Today's scientific high-performance computing applications and advanced instruments are producing vast volumes of data across a wide range of domains, which impose a serious burden on data transfer and storage. Error-bounded lossy compression has been developed and widely used in the scientific community because it not only can significantly reduce the data volumes but also can strictly control the data distortion based on the user-specified error bound. Existing lossy compressors, however, cannot offer ultrafast compression speed, which is highly demanded by numerous applications or use cases (such as in-memory compression and online instrument data compression). In this paper, we propose a novel ultrafast error-bounded lossy compressor that can obtain fairly high compression performance on both CPUs and GPUs and with reasonably high compression ratios. The key contributions are threefold. (1) We propose a generic error-bounded lossy compression framework---called SZx---that achieves ultrafast performance through its novel design comprising only lightweight operations such as bitwise and addition/subtraction operations, while still keeping a high compression ratio. (2) We implement SZx on both CPUs and GPUs and optimize the performance according to their architectures. (3) We perform a comprehensive evaluation with six real-world production-level scientific datasets on both CPUs and GPUs. Experiments show that SZx is 2~16x faster than the second-fastest existing error-bounded lossy compressor (either SZ or ZFP) on CPUs and GPUs, with respect to both compression and decompression.

Yu, Xiaodong↗
DOE OSTI2018

Overcoming fatigue through compression for advanced elastocaloric cooling

Elastocaloric materials exhibit extraordinary cooling potential, but the repetition of cyclic mechanical loadings during long-term operation of cooling systems requires the refrigerant material to have long fatigue life. Here, this paper reviews the fundamental cause of fatigue from aspects of initiation and propagation of fatigue cracks in shape-memory alloys (SMAs) that are used as elastocaloric materials, and highlights recent advances in using compression to overcome fatigue by curtailing the generation of surfaces associated with crack propagation. Compression is identified as a key means to extend fatigue lifetime in engineering design of elastocaloric cooling drive mechanisms. We summarize the state-of-the-art performance of different SMAs as elastocaloric materials and discuss the influence of low cyclic strains and high resistance to transformation. We present integration of compression-based material assemblies into a cooling system prototype and optimization of the system efficiency using work recovery and related measures.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Reinforcement Learning as a Parsimonious Alternative to Prediction Cascades: A Case Study on Image Segmentation

Deep learning architectures have achieved state-of-the-art (SOTA) performance on computer vision tasks such as object detection and image segmentation. This may be attributed to the use of over-parameterized, monolithic deep learning architectures executed on large datasets. Although such large architectures lead to increased accuracy, this is usually accompanied by a larger increase in computation and memory requirements during inference. While this is a non-issue in traditional machine learning (ML) pipelines, the recent confluence of machine learning and fields like the Internet of Things (IoT) has rendered such large architectures infeasible for execution in low-resource settings. For some datasets, large monolithic pipelines may be overkill for simpler inputs. To address this problem, previous efforts have proposed decision cascades where inputs are passed through models of increasing complexity until the desired performance is achieved. However, we argue that cascaded prediction leads to sub-optimal throughput and increased computational cost due to wasteful intermediate computations. To address this, we propose PaSeR (Parsimonious Segmentation with Reinforcement Learning) a non-cascading, cost-aware learning pipeline as an efficient alternative to cascaded decision architectures. Through experimental evaluation on both real-world and standard datasets, we demonstrate that PaSeR achieves better accuracy while minimizing computational cost relative to cascaded models. Further, we introduce a new metric IoU/GigaFlop to evaluate the balance between cost and performance. On the real-world task of battery material phase segmentation, PaSeR yields 179% improvement over SOTA MatPhase model and a 196% improvement over IDK Cascades under the IoU/GigaFlop metric. We also demonstrate PaSeR’s adaptability to complementary models trained on a noisy MNIST dataset, where it outperforms all baselines on IoU/GigaFlop by an average of 44%.

Srikshan, Bharat↗
DOE OSTI2021

Model fusion with physics-guided machine learning: Projection-based reduced-order modeling

The unprecedented amount of data generated from experiments, field observations, and large-scale numerical simulations at a wide range of spatiotemporal scales has enabled the rapid advancement of data-driven and especially deep learning models in the field of fluid mechanics. Although these methods are proven successful for many applications, there is a grand challenge of improving their generalizability. This is particularly essential when data-driven models are employed within outer-loop applications like optimization. In this work, we put forth a physics-guided machine learning (PGML) framework that leverages the interpretable physics-based model with a deep learning model. Leveraging a concatenated neural network design from multi-modal data sources, the PGML framework is capable of enhancing the generalizability of data-driven models and effectively protects against or inform about the inaccurate predictions resulting from extrapolation. We apply the PGML framework as a novel model fusion approach combining the physics-based Galerkin projection model and long- to short-term memory (LSTM) network for parametric model order reduction of fluid flows. We demonstrate the improved generalizability of the PGML framework against a purely data-driven approach through the injection of physics features into intermediate LSTM layers. Our quantitative analysis shows that the overall model uncertainty can be reduced through the PGML approach, especially for test data coming from a distribution different than the training data. Moreover, we demonstrate that our approach can be used as an inverse diagnostic tool providing a confidence score associated with models and observations. The proposed framework also allows for multi-fidelity computing by making use of low-fidelity models in the online deployment of quantified data-driven models.

42 ENGINEERING↗
DOE OSTI2026

Phase Synergy Enables Low‐Power Ferroelectric Switching in HfO 2 Epitaxial Films

HfO 2 -based ferroelectric materials have emerged as leading candidates for next-generation non-volatile memory technologies, owing to their nanoscale robust ferroelectricity and complementary metal–oxide–semiconductor (CMOS) compatibility. However, challenges and debates persist in advancing and comprehensively understanding their ferroelectric behavior. In particular, conventional approaches typically regard non-ferroelectric phases as detrimental and primarily focus on suppressing their formation, yet overlooking their potentially synergistic contributions—particularly those of the tetragonal (T) phase. Here, we unambiguously clarify the beneficial role of the T-phase and introduce a phase-boundary engineering strategy that deliberately harnesses it to enhance ferroelectricity in HfO 2 films. By stabilizing optimal coherent boundaries between ferroelectric orthorhombic (O) and T phases in epitaxial La-doped HfO 2 films, we achieve significant improvements in ferroelectric properties—doubling the remanent polarization (P r ∼ 30 µC/cm 2 ) and substantially reducing the coercive field (E c ∼ 3 MV/cm) by 30% compared to low-La doped samples without such boundaries. Atomic-scale electron microscopy reveals the structural nature of the atomically sharp, coherent O–T boundaries. Combined with deep-learning enhanced molecular dynamics simulations, our results unravel that these boundaries facilitate intermediate polarization states that lower the switching energy barrier. Consequently, phase coexistence shifts from an inherent drawback to a tunable design element, offering a broadly applicable route to ultra-low-power HfO 2 -based nanoelectronics.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2017

Generation of feasible integer solutions on a massively parallel computer using the feasibility pump

Here, we present an approach to parallelize generation of feasible mixed integer solutions of mixed integerlinear programs in distributed memory high performance computing environments. This approachcombines a parallel framework with feasibility pump (FP) as the rounding heuristic. It runs multiple FPinstances with different starting solutions concurrently, while allowing them to share information. Ourcomputational results suggest that the improvement resulting from parallelization using our approach is statistically significant.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Surrogate models for development of unconventional shale reservoirs by an integrated numerical approach of hydraulic fracturing, flow and geomechanics, and machine learning

We develop well-completion surrogate models by taking an integrated workflow of hydraulic fracturing, flow, geomechanics, and machine learning simulation. There are three steps in the proposed workflow. First, history-matching processes are conducted with the field data including pumping and production data for characterization. Second, full-physics simulation is performed with various parameters of the field development (e.g., cluster spacing, clusters per stage, pumping rates and times, amount of proppant, and well spacing) to generate multiple simulation results by changing the parameters of the completion design with well-known hydraulic fracturing, reservoir, geomechanics simulators to calculate fracture geometry, reservoir depressurization, induced stress changes. The workflow is demonstrated over a field in the Southern Midland Basin. Here, we take two completion scenarios: a single well case followed by a multi-well case. Finally, a Long Short-Term Memory (LSTM) machine learning algorithm is employed to create surrogate models that can replicate the full-physics simulation results. Furthermore, results show that the trained models applied in the single well and multi-well cases for a particular geological system can provide good accuracy close to those provided by full-physics simulations. Specifically, the site-specific surrogate models can predict fracture parameters (length, height, and surface area) and cumulative production accurately with computational efficiency, suggesting our proposed workflow can be used as a pragmatic tool for expediting the well completion optimization process.

Geomechanics↗
DOE OSTI2017

UNITY: Unified Memory and File Space

This paper describes the vision for UNITY, a new high-performance computing focused data storage abstraction that places the entire memory hierarchy, including both traditionally separated memoryand file-based data storage, into one storage continuum. Through the use of a novel API and a set of services centered around a smart runtime system, UNITY is able to provide a number of valuable and interesting benefits. The unified storage space provides a scalable and resilient data environment that dynamically manages the mapping of data onto available resources based on multiple factors, including desired persistence and energy budget considerations. By eliminating the need for high-performance computing domain scientists to develop architecture-dependent optimizations for rapidly evolving data storage technologies, UNITY addresses both ease-of-use and performance.

Otstott, Douglas↗
DOE OSTI2020

Interfacial spin–orbit torques

Spin–orbit torques offer a promising mechanism for electrically controlling magnetization dynamics in nanoscale heterostructures. While spin–orbit torques occur predominately at interfaces, the physical mechanisms underlying these torques can originate in both the bulk layers and at interfaces. Classifying spin–orbit torques based on the region that they originate in provides clues as to how to optimize the effect. While most bulk spin–orbit torque contributions are well studied, many of the interfacial contributions allowed by symmetry have yet to be fully explored theoretically and experimentally. To facilitate progress, we review interfacial spin–orbit torques from a semiclassical viewpoint and relate these contributions to recent experimental results. In this study, we show the relationship between different interface transport parameters within the same model. For charges and spins flowing perpendicular to the interface, interfacial spin–orbit coupling both modifies the mixing conductance of the magnetoelectronic circuit theory and gives rise to spin memory loss. For in-plane electric fields, interfacial spin–orbit coupling gives rise to torques described by spin–orbit filtering, spin swapping, and precession. In addition, these same interfacial processes generate spin currents that flow into the non-magnetic layer. For in-plane electric fields in trilayer structures, the spin currents generated at the interface between one ferromagnetic layer and the non-magnetic spacer layer can propagate through the non-magnetic layer to produce novel torques on the other ferromagnetic layer.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2024

Geometry-aware training of factorized layers in tensor Tucker format

Reducing parameter redundancies in neural network architectures is crucial for achieving feasible computational and memory requirements during train and inference of large networks. Given its easy implementation and flexibility, one promising approach is layer factorization, which reshapes weight tensors into a matrix format and parameterizes it as the product of two rank-r matrices. However, this family of approaches often requires an initial full-model warm-up phase, prior knowledge of a feasible rank, and it is sensitive to parameter initialization.In this work, we introduce a novel approach to train the factors of a Tucker decomposition of the weight tensors. Our training proposal proves to be optimal in locally approximating the original unfactorized dynamics and stable for the initialization. Furthermore, the rank of each mode is dynamically updated during training.We provide a theoretical analysis of the algorithm, showing convergence, approximation and local descent guarantees. The method's performance is further illustrated through a variety of experiments, showing remarkable training compression rates and comparable or even better performance than the full baseline and alternative layer factorization strategies.

Zangrando, Emanuele [Gran Sasso Science Institute ↗
DOE OSTI2025

Distributed Stochastic Optimization of a Neural Representation Network for Time-Space Tomography Reconstruction

4D time-space reconstruction of dynamic events or deforming objects using X-ray computed tomography (CT) is an important inverse problem in non-destructive evaluation. Conventional back-projection based reconstruction methods assume that the object remains static for the duration of several tens or hundreds of X-ray projection measurement images (reconstruction of consecutive limited-angle CT scans). However, this is an unrealistic assumption for many in-situ experiments that causes spurious artifacts and inaccurate morphological reconstructions of the object. To solve this problem, we propose to perform a 4D time-space reconstruction using a distributed implicit neural representation (DINR) network that is trained using a novel distributed stochastic training algorithm. Our DINR network learns to reconstruct the object at its output by iterative optimization of its network parameters such that the measured projection images best match the output of the CT forward measurement model. Here, we use a forward measurement model that is a function of the DINR outputs at a sparsely sampled set of continuous valued 4D object coordinates. Unlike previous neural representation architectures that forward and back propagate through dense voxel grids that sample the object's entire time-space coordinates, we only propagate through the DINR at a small subset of object coordinates in each iteration resulting in an order-of-magnitude reduction in memory and compute for training. DINR leverages distributed computation across several compute nodes and GPUs to produce high-fidelity 4D time-space reconstructions. We use both simulated parallel-beam and experimental cone-beam X-ray CT datasets to demonstrate the superior performance of our approach.

36 MATERIALS SCIENCE↗
DOE OSTI2022

Towards Enhancing Coding Productivity for GPU Programming Using Static Graphs

The main contribution of this work is to increase the coding productivity of GPU programming by using the concept of Static Graphs. GPU capabilities have been increasing significantly in terms of performance and memory capacity. However, there are still some problems in terms of scalability and limitations to the amount of work that a GPU can perform at a time. To minimize the overhead associated with the launch of GPU kernels, as well as to maximize the use of GPU capacity, we have combined the new CUDA Graph API with the CUDA programming model (including CUDA math libraries) and the OpenACC programming model. We use as test cases two different, well-known and widely used problems in HPC and AI: the Conjugate Gradient method and the Particle Swarm Optimization. In the first test case (Conjugate Gradient) we focus on the integration of Static Graphs with CUDA. In this case, we are able to significantly outperform the NVIDIA reference code, reaching an acceleration of up to 11x thanks to a better implementation, which can benefit from the new CUDA Graph capabilities. In the second test case (Particle Swarm Optimization), we complement the OpenACC functionality with the use of CUDA Graph, achieving again accelerations of up to one order of magnitude, with average speedups ranging from 2x to 4x, and performance very close to a reference and optimized CUDA code. Our main target is to achieve a higher coding productivity model for GPU programming by using Static Graphs, which provides, in a very transparent way, a better exploitation of the GPU capacity. The combination of using Static Graphs with two of the current most important GPU programming models (CUDA and OpenACC) is able to reduce considerably the execution time w.r.t. the use of CUDA and OpenACC only, achieving accelerations of up to more than one order of magnitude. Finally, we propose an interface to incorporate the concept of Static Graphs into the OpenACC Specifications.

58 GEOSCIENCES↗
DOE OSTI2014

Hierarchical computation in the SPMD programming model

Large-scale parallel machines are programmed mainly with the single program, multiple data (SPMD) model of parallelism. While this model has advantages of scalability and simplicity, it does not fit well with divide-and-conquer parallelism or hierarchical machines that mix shared and distributed memory. In this paper, we define the recursive single program, multiple data model (RSPMD) that extends SPMD with a hierarchical team mechanism to support hierarchical algorithms and machines. We implement this model in the Titanium language and describe how to eliminate a class of deadlocks by ensuring alignment of collective operations. We present application case studies evaluating the RSPMD model, showing that it enables divide-and-conquer algorithms such as sorting to be elegantly expressed and that team collective operations increase performance of conjugate gradient by up to a factor of two. The model also facilitates optimizations for hierarchical machines, improving scalability of particle in cell by 8x and performance of sorting and a stencil code by up to 40% and 14%, respectively.

Kamil, A↗
DOE OSTI2026

A cell-centered AMR-ALE framework for 3D multi-material hydrodynamics. Part I: Lagrangian and indirect Euler AMR algorithms

Many applications of physics and engineering involve wide ranges of time and spatial scales. The numerical simulation of localized small scales such as shock waves and material interfaces requires a large number of computational cells in these regions. For these applications, Lagrangian and Arbitrary-Lagrangian-Eulerian (ALE) related methods are engaging since the moving mesh feature naturally brings mesh cells on shock discontinuities and material interfaces are carefully captured. In addition, Adaptive-Mesh-Refinement (AMR) strategies aim to optimize computational resources by concentrating finer mesh cells only in areas of interest while using coarser cells elsewhere. A key but challenging AMR requirement consists in efficiently distributing the computational effort to achieve high accuracy without the prohibitive computational costs associated with uniformly fine grids. Here, in this document, the coupling of the p4est AMR library with a cell-centered Lagrangian scheme is presented with the goal to perform reliable 3D Lagrangian-AMR and indirect Euler-AMR multi-material simulations. In particular, it is shown that starting from a 3D indirect ALE code, the memory management and load balancing requirements can be delegated to an external library (here the p4est library) to unlock ALE-AMR capabilities. First, we present a strategy to transcribe the octant-based connectivity of the 3D AMR framework with that of an unstructured mesh of polygonal cells used in Lagrangian hydrodynamics. Then, we show how refinement and coarsening operations must be adapted to the particular Lagrangian framework to ensure the conservation of volume during those steps. Finally, several numerical test cases are presented that demonstrate the capabilities of the Lagrangian-AMR and indirect Euler-AMR algorithms.

3D cell-centered Lagrangian numerical scheme↗
DOE OSTI2016

An efficient and numerically stable procedure for generating sextic force fields in normal mode coordinates

In this paper, we outline a general, scalable, and black-box approach for calculating high-order strongly coupled force fields in rectilinear normal mode coordinates, based upon constructing low order expansions in curvilinear coordinates with naturally limited mode-mode coupling, and then transforming between coordinate sets analytically. The optimal balance between accuracy and efficiency is achieved by transforming from 3 mode representation quartic force fields in curvilinear normal mode coordinates to 4 mode representation sextic force fields in rectilinear normal modes. Using this reduced mode-representation strategy introduces an error of only 1 cm{sup −1} in fundamental frequencies, on average, across a sizable test set of molecules. We demonstrate that if it is feasible to generate an initial semi-quartic force field in curvilinear normal mode coordinates from ab initio data, then the subsequent coordinate transformation procedure will be relatively fast with modest memory demands. This procedure facilitates solving the nuclear vibrational problem, as all required integrals can be evaluated analytically. Our coordinate transformation code is implemented within the extensible PyPES library program package, at http://sourceforge.net/projects/pypes-lib-ext/.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗