Search NASA⌕ Search

SEARCH · Search NASA

Results for “Performance analysis”

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 1,009 records · Page 56

Vacuum hydride epitaxy of silicon: kinetics of monosilane pyrolysis on the growth surface

Analytical expressions relating the rate of silicon atom incorporation into a growing crystal to the typical frequency of silane molecule pyrolysis on the silicon surface in the growth temperature range are derived. Based on currently available experimental data, the range of typical decomposition frequencies of hydride molecule radicals adsorbed at the silicon wafer surface in the temperature range of 450-700 Degree-Sign C is determined for the most widely used physicochemical models. It is shown that the most probable molecular decomposition model can be chosen based on the experimental study of the temperature dependence of the decomposition rate of adsorbed hydride molecules. A change in the silane molecule pyrolysis rate or the hydrogen desorption rate from the surface in principle makes it possible to increase the Si layer growth rate without additional substrate heating under conditions of low-temperature epitaxy (450-550 Degree-Sign C), but no larger than by a factor of 2-3 in the former case and up to 100 times in the latter case. The analysis performed shows that physicochemical pyrolysis models in which hydrogen is trapped by the surface, mostly at the stage of decomposition of silane radicals adsorbed by the surface, are more realistic.

36 MATERIALS SCIENCE↗

Stochastic Reductions for Inertial Fluid-Structure Interactions Subject to Thermal Fluctuations

We explore the dynamics of elastic microstructures within a fluid that are subjected to thermal fluctuations. We perform analysis to obtain systematically simplified descriptions of the mechanics in the limiting regimes when (i) the coupling forces that transfer momentum between the fluid and microstructures are strong, (ii) the mass of the microstructures is small relative to the displaced mass of the fluid, and (iii) the response to stresses results in hydrodynamics that relax rapidly to a quasi-steady-state relative to the motions of the microstructure. Furthermore, we derive effective equations using a singular perturbation analysis of the backward Kolmogorov equations of the stochastic process. Our continuum mechanics description is based on the stochastic Eulerian--Lagrangian method which provides a framework for approximation of the fluid-structure interactions when subject to thermal fluctuations.

97 MATHEMATICS AND COMPUTING↗

Towards Batched Linear Solvers on Accelerated Hardware Platforms

As hardware evolves, an increasingly effective approach to develop energy efficient, high-performance solvers, is to design them to work on many small and independent problems. Indeed, many applications already need this functionality, especially for GPUs, which are known to be currently about four to five times more energy efficient than multicore CPUs for every floating-point operation. In this paper, we describe the development of the main one-sided factorizations: LU, QR, and Cholesky; that are needed for a set of small dense matrices to work in parallel. We refer to such algorithms as batched factorizations. Our approach is based on representing the algorithms as a sequence of batched BLAS routines for GPU-contained execution. Note that this is similar in functionality to the LAPACK and the hybrid MAGMA algorithms for large-matrix factorizations. But it is different from a straightforward approach, whereby each of GPU's symmetric multiprocessors factorizes a single problem at a time. We illustrate how our performance analysis together with the profiling and tracing tools guided the development of batched factorizations to achieve up to 2-fold speedup and 3-fold better energy efficiency compared to our highly optimized batched CPU implementations based on the MKL library on a two-sockets, Intel Sandy Bridge server. Compared to a batched LU factorization featured in the NVIDIA's CUBLAS library for GPUs, we achieves up to 2.5-fold speedup on the K40 GPU.

Haidar, Azzam↗

Progress Towards Accelerating the Unified Model on Hybrid Multi-Core Systems

The cloud microphysics scheme, CASIM, and the radiation scheme, SOCRATES, are two computationally intensive parts within the Met Office's Unified Model (UM). This study enables CASIM and SOCRATES to use accelerated multi-core systems for optimal computational performance of the UM. Using profiling to guide our efforts, we refactored the code for optimal threading and kernel arrangement and implemented OpenACC directives manually or through the CLAW source-to-source translator. Initial porting results achieved 10.02x and 9.25x speedup in CASIM and SOCRATES respectively on 1 GPU compared with 1 CPU core. A granular performance analysis of the strategy and bottlenecks are discussed. These improvements will enable UM to run on heterogeneous computers and a path forward for further improvements is provided.

Zhang, Wei↗

Accelerating Random Forest Classification on GPU and FPGA

Random Forests (RFs) are a commonly used machine learning method for classification and regression tasks spanning a variety of application domains, including bioinformatics, business analytics, and software optimization. While prior work has focused primarily on improving performance of the training of RFs, many applications, such as malware identification, cancer prediction, and banking fraud detection, require fast RF classification. In this work, we accelerate RF classification on GPU and FPGA. In order to provide efficient support for large datasets, we propose a hierarchical memory layout suitable to the GPU/FPGA memory hierarchy. We design three RF classification code variants based on that layout, and we investigate GPU- and FPGA-specific considerations for these kernels. Our experimental evaluation, performed on an Nvidia Xp GPU and on a Xilinx Alveo U250 FPGA accelerator card using publicly available datasets on the scale of millions of samples and tens of features, covers various aspects. First, we evaluate the performance benefits of our hierarchical data structure over the standard compressed sparse row (CSR) format. Second, we compare our GPU implementation with cuML, a machine learning library targeting Nvidia GPUs. Third, we explore the performance/accuracy tradeoff resulting from the use of different tree depths in the RF. Finally, we perform a comparative performance analysis of our GPU and FPGA implementations. Our evaluation shows that for high accuracy targets, our GPU implementation yields 5-9x speedup over CSR, and up to a 2x speedup over cuML.

FPGA, Xilinx FPGA, GPU, Random Forest classificati↗

A Bayesian Approach for Quantifying Data Scarcity when Modeling Human Behavior via Inverse Reinforcement Learning

Computational models that formalize complex human behaviors enable study and understanding of such behaviors. However, collecting behavior data required to estimate the parameters of such models is often tedious and resource intensive. Thus, estimating dataset size as part of data collection planning (also known as Sample Size Determination) is important to reduce the time and effort of behavior data collection while maintaining an accurate estimate of model parameters. In this paper, we present a sample size determination method based on Uncertainty Quantification (UQ) for a specific Inverse Reinforcement Learning (IRL) model of human behavior, in two cases: 1) pre-hoc experiment design—conducted in the planning stage before any data is collected, to guide the estimation of how many samples to collect; and 2) post-hoc dataset analysis—performed after data is collected, to decide if the existing dataset has sufficient samples and whether more data is needed. Here, we validate our approach in experiments with a realistic model of behaviors of people with Multiple Sclerosis (MS) and illustrate how to pick a reasonable sample size target. Our work enables model designers to perform a deeper, principled investigation of effects of dataset size on IRL.

97 MATHEMATICS AND COMPUTING↗

Scaling Resolution of Gigapixel Whole Slide Images Using Spatial Decomposition on Convolutional Neural Networks

Gigapixel images are prevalent in scientific domains ranging from remote sensing, and satellite imagery to microscopy, etc. However, training a deep learning model at the natural resolution of those images has been a challenge in terms of both, overcoming the resource limit (e.g. HBM memory constraints), as well as scaling up to a large number of GPUs. In this paper, we trained Residual neural Networks (ResNet) on 22,528 x 22,528-pixel size images using a distributed spatial decomposition method on 2,304 GPUs on the Summit Supercomputer. We applied our method on a Whole Slide Imaging (WSI) dataset from The Cancer Genome Atlas (TCGA) database. WSI images can be in the size of 100,000 x 100,000 pixels or even larger, and in this work we studied the effect of image resolution on a classification task, while achieving state-of-the-art AUC scores. Moreover, our approach doesn't need pixel-level labels, since we're avoiding patching from the WSI images completely, while adding the capability of training arbitrary large-size images. This is achieved through a distributed spatial decomposition method, by leveraging the non-block fat-tree interconnect network of the Summit architecture, which enabled GPU-to-GPU direct communication. Finally, detailed performance analysis results are shown, as well as a comparison with a data-parallel approach when possible.

Tsaris, Aristeidis (aris)↗

BCS (Beamline Control System) v2

The software is a flexible architecture used to integrate motors, analog data, digital signals, and instruments together to control scientific experiments and collect data during an experiment. The architecture supports features needed by scientists to quickly add new hardware to an experiment and to take and view data in a meaningful way. It does not perform analysis of the data, but provides basic visualization needed to evaluate the quality of the data as well as the calibration of the system.

Cornell, Earl↗

Adiak

Adiak is a computer science data management interface tool for collecting metadata about an application's runs and disseminating it to tools that subscribe to this data. Examples of this metadata could be what user is running an application, what systems it runs on, or application provided name/value pairs. Adiak's capabilities include automated reading of system-level metadata, and a generic type system for applications to register their own name/value metadata. Adiak is currently designed to provide a standard interface that can be integrated with MPI-based applications that run on high performance computing systems. Example tools that use Adiak metadata include performance analysis tools and workflow tools. By allowing the application to specify metadata to Adiak, which then disseminates it to tools, application developers do not have to write to tool-specific interfaces.

LeGendre, MatthewP↗

JEGA V 2.3

JEGA is a software package that implements a multi-objective genetic algorithm (MOGA) for solution to multi-objective optimization problems (MOPS) and a single objective genetic algorithm (SOGA) for solution to single objective optimization problems (SOPS). JEGA is highly configurable and provides a rich set of algorithmic components for tailored optimization. It has been used successfully on many problems by users both internal and external to Sandia. By employing object-oriented design to implement abstractions of the key components making up the algorithms, JEGA provides a flexible and extensible problem-solving environment for design and performance analysis of computational models.

EDDY, JOHN↗

pySimFrac

pySimFrac is a python library for generating 3D synthetic fracture realizations, integrating with fluid simulators, and performing analysis.

Guiltinan, Eric↗

DISTRI: Distributed Multi-Facility HPC Simulator (DISTRI) v2.1

DISTRI is an advanced network simulator designed for multi-facility computational infrastructures with agentic behavior. It simulates HPC facilities where computational resources act as autonomous agents, making intelligent decisions about job scheduling, load balancing, and resource allocation. The simulator focuses on developing and testing decentralized algorithms that promote resilience and efficiency in multi-facility environments. Key Features: - Agentic Resource Behavior: Processors and DTNs act as autonomous agents with decision-making capabilities - Pheromone-Based Load Balancing: Decentralized load balancing inspired by ant colony optimization - Dual Topology Support: Mesh (normal operations) and Dumbell (network testing) topologies - Comprehensive TCP Simulation: Realistic TCP implementations with multiple congestion control algorithms - Failure Resilience Testing: Processor failure simulation with automatic job reassignment - Extensive Visualization: Detailed performance analysis and metrics collection - Research-Ready: Designed for algorithm development and benchmarking

Bez, Jean Luca [Lawrence Berkeley National Laborat↗

cabbi-bio/sorghum-sugarcane-RBCS-RAF1-2024

This repository includes data sets and R scripts that were used to perform analysis and produce figures for the following publication: Salesse-Smith, C. E. et al. "Adapting C4 photosynthesis to atmospheric change and increasing productivity by elevating Rubisco content in sorghum and sugarcane." Proceedings of the National Academy of Sciences 122, e2419943122 (2025) doi:10.1073/pnas.2419943122.

Salesse-Smith, Coralie↗

Estimation of Arrivals on Green at Signalized Intersections Using Stop-Bar Video Detection

Across the world, traffic congestion is increasing with alarming rapidity. Traffic signal control effectiveness, in coordinated networks, is often investigated in relation to the type of vehicle arrivals at the signalized intersections. Recently, several transportation agencies have switched from traditional loop detectors to video detection. When video cameras are accompanied by computer vision, one can extract more information about traffic “dynamics” than by using traditional inductive loop detectors. Collecting arrival times of multiple vehicles after the first arrival at the stop-bar detector might be challenging when using inductive loop detectors (since after the first arrival, detector status is always occupied). However, emerging video detection systems allow tracking of each vehicle’s entrance time in the detection zone, departure time from the detection zone, and the type of vehicle. This information can be used to estimate vehicular arrival and departure times, which then can be fed into machine learning algorithms to estimate arrivals on green (AOG). However, such research ideas have not been documented so far. Thus, this paper presents an estimation model for AOG, which was developed using multigene genetic programming. A robust experimental dataset was collected from a highly calibrated and validated microsimulation model of an 11-intersection corridor in Chattanooga, TN. The results of the model’s performance analysis showed the high accuracy of the training-, testing-, and validation datasets. The practical benefit of this model is that it can be applied to estimate arrival types at intersections where only stop-bar video detection exists.

Engineering↗

Batched matrix computations on hardware accelerators based on GPUs

Scientific applications require solvers that work on many small size problems that are independent from each other. At the same time, the high-end hardware evolves rapidly and becomes ever more throughput-oriented and thus there is an increasing need for an effective approach to develop energy-efficient, high-performance codes for these small matrix problems that we call batched factorizations. The many applications that need this functionality could especially benefit from the use of GPUs, which currently are four to five times more energy efficient than multicore CPUs on important scientific workloads. This study, consequently, describes the development of the most common, one-sided factorizations, Cholesky, LU, and QR, for a set of small dense matrices. The algorithms we present together with their implementations are, by design, inherently parallel. In particular, our approach is based on representing the process as a sequence of batched BLAS routines that are executed entirely on a GPU. Importantly, this is unlike the LAPACK and the hybrid MAGMA factorization algorithms that work under drastically different assumptions of hardware design and efficiency of execution of the various computational kernels involved in the implementation. Thus, our approach is more efficient than what works for a combination of multicore CPUs and GPUs for the problems sizes of interest of the application use cases. The paradigm where upon a single chip (a GPU or a CPU) factorizes a single problem at a time is not at all efficient in our applications’ context. We illustrate all of these claims through a detailed performance analysis. With the help of profiling and tracing tools, we guide our development of batched factorizations to achieve up to two-fold speedup and three-fold better energy efficiency as compared against our highly optimized batched CPU implementations based on MKL library. Finally, the tested system featured two sockets of Intel Sandy Bridge CPUs and we compared with a batched LU factorizations featured in the CUBLAS library for GPUs, we achieve as high as 2.5× speedup on the NVIDIA K40 GPU.

97 MATHEMATICS AND COMPUTING↗

Data for Adapting C4 Photosynthesis to Atmospheric Change and Increasing Productivity by Elevating Rubisco Content in Sorghum and Sugarcane

This repository includes data sets and R scripts that were used to perform analysis and produce figures for the following publication: Salesse-Smith, C. E. et al. “Adapting C4 photosynthesis to atmospheric change and increasing productivity by elevating Rubisco content in sorghum and sugarcane.” Proceedings of the National Academy of Sciences 122, e2419943122 (2025) doi:10.1073/pnas.2419943122.

Biomass Analytics↗

Molecular simulation data for 'Data-guided Multi-Map variables for ensemble refinement of molecular movies'

These trajectories, scripts, and analysis performed on Summit underly the work published as 'Data-guided Multi-Map variables for ensemble refinement of molecular movies'. The trajectories include equilibrium and non-equilibrium sampling of ADK, CODH, and FLPP3, the scripts used to build the systems, and the scripts used to analyze the output. The directory structure is explained further in an internal README file.

59 BASIC BIOLOGICAL SCIENCES↗