Search NASA⌕ Search

SEARCH · Search NASA

Results for “Kernel Designs”

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 37 records · Page 2
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

A neural network‐enhanced reproducing kernel particle method for modeling strain localization

Abstract Modeling the localized intensive deformation in a damaged solid requires highly refined discretization for accurate prediction, which significantly increases the computational cost. Although adaptive model refinement can be employed for enhanced effectiveness, it is cumbersome for the traditional mesh‐based methods to perform while modeling the evolving localizations. In this work, neural network‐enhanced reproducing kernel particle method (NN‐RKPM) is proposed, where the location, orientation, and shape of the solution transition near a localization is automatically captured by the NN approximation via a block‐level neural network (NN) optimization. The weights and biases in the blocked parameterization network control the location and orientation of the localization. The designed basic four‐kernel NN block is capable of capturing a triple junction or a quadruple junction topological pattern, while more complicated localization topological patters are captured by the superposition of multiple four‐kernel NN blocks. The standard RK approximation is then utilized to approximate the smooth part of the solution, which permits a much coarser discretization than the high‐resolution discretization needed to capture sharp solution transitions with the conventional methods. A regularization of the NN approximation is additionally introduced for discretization‐independent material responses. The effectiveness of the proposed NN‐RKPM is verified by a series of numerical verifications.

Baek, Jonghyuk↗
DOE OSTI2023

Kernel Buffer Volume Fraction Margin of the AGR Designed Fuel Particle

Modeling results used to assess the fuel performance of the TRISO-coated fuel particles as a function of kernel/buffer volume fraction include SiC tangential stress, formation of the buffer/IPyC gap, particle temperature profile, internal particle pressure, fission gas released from the kernel, probability of fuel particle failure, and fission product diffusion. These results were evaluated at two burnup levels and irradiation temperatures to bound expected steady-state irradiation conditions. In general, increasing the kernel/buffer volume fraction increases the SiC stress and subsequently the failure probability of a fuel particle when compared to the AGR designed particle. There was little impact on the fission product diffusion through the particle as the kernel/buffer volume fraction increased.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2025

Working with Bézier Curves as bases for Functional Expansion Tallies

Functional expansion tallies (FETs) are powerful tools for getting more information per history from Monte Carlo simulations, but in the past they have been constrained to orthogonal bases. Bézier curves are used widely in computer aided design (CAD) geometry kernels and could be well-suited for FETs due to their ability to assume many arbitrary shapes, but they use nonorthogonal bases. Recent developments in 2021 have made nonorthogonal FETs possible. The convergence of Bézier curve FETs in both polynomial order and number of samples is explored in this work. It is shown that these bases are well-suited for representing normal distributions and this opens the door to the possibility of other CAD-derived FET bases.

97 - MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Hierarchical Speed Planner for Automated Vehicles: A Framework for Lagrangian Variable Speed Limit in Mixed-Autonomy Traffic

Here, this article presents a novel hierarchical speed planning framework for variable speed limits in mixed-autonomy traffic environments, leveraging server-side macroscopic control and vehicle-side microscopic execution. The framework integrates real-time traffic state estimation (TSE) and reinforcement learning (RL)-based control to mitigate congestion and improve traffic flow. A TSE enhancement module combines macroscopic data from sources like INRIX with high-resolution observations from connected autonomous vehicles (CAVs), enabling predictive modeling to address latency and noise. The target speed design module employs kernel smoothing and a buffer zone strategy to optimize traffic density and flow around bottlenecks. The proposed system was validated in the largest open-road test to date with 100 CAVs, demonstrating an overall 8% traffic density decrease, with a specific decrease of 7% upstream, 10% downstream, and a 52% decrease during the congestion formation phase at bottlenecks.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2025

Physics-Informed Gaussian Process Inference of Liquid Structure from Scattering Data

We present a nonparametric Bayesian framework to infer radial distribution functions from experimental scattering measurements with uncertainty quantification using nonstationary Gaussian processes. The Gaussian process prior mean and kernel functions are designed to mitigate well-known numerical challenges with the Fourier transform, including discrete measurement binning and detector windowing, while encoding fundamental yet minimal physical knowledge of the liquid structure. We demonstrate uncertainty propagation of the Gaussian process posterior to unmeasured quantities of interest. Experimental radial distribution functions of liquid argon and water with uncertainty quantification are provided as both a proof of principle for the method and a benchmark for molecular models.

Chemical structure↗
NASA NTRS2002

Design and Analysis of Architectures for Structural Health Monitoring Systems

During the two-year project period, we have worked on several aspects of Health Usage and Monitoring Systems for structural health monitoring. In particular, we have made contributions in the following areas. 1. Reference HUMS architecture: We developed a high-level architecture for health monitoring and usage systems (HUMS). The proposed reference architecture is shown. It is compatible with the Generic Open Architecture (GOA) proposed as a standard for avionics systems. 2. HUMS kernel: One of the critical layers of HUMS reference architecture is the HUMS kernel. We developed a detailed design of a kernel to implement the high level architecture.3. Prototype implementation of HUMS kernel: We have implemented a preliminary version of the HUMS kernel on a Unix platform.We have implemented both a centralized system version and a distributed version. 4. SCRAMNet and HUMS: SCRAMNet (Shared Common Random Access Memory Network) is a system that is found to be suitable to implement HUMS. For this reason, we have conducted a simulation study to determine its stability in handling the input data rates in HUMS. 5. Architectural specification.

Mukkamala, Ravi↗
NASA NTRS2014

Collaborative WorkBench for Researchers - Work Smarter, Not Harder

It is important to define some commonly used terminology related to collaboration to facilitate clarity in later discussions. We define provisioning as infrastructure capabilities such as computation, storage, data, and tools provided by some agency or similarly trusted institution. Sharing is defined as the process of exchanging data, programs, and knowledge among individuals (often strangers) and groups. Collaboration is a specialized case of sharing. In collaboration, sharing with others (usually known colleagues) is done in pursuit of a common scientific goal or objective. Collaboration entails more dynamic and frequent interactions and can occur at different speeds. Synchronous collaboration occurs in real time such as editing a shared document on the fly, chatting, video conference, etc., and typically requires a peer-to-peer connection. Asynchronous collaboration is episodic in nature based on a push-pull model. Examples of asynchronous collaboration include email exchanges, blogging, repositories, etc. The purpose of a workbench is to provide a customizable framework for different applications. Since the workbench will be common to all the customized tools, it promotes building modular functionality that can be used and reused by multiple tools. The objective of our Collaborative Workbench (CWB) is thus to create such an open and extensible framework for the Earth Science community via a set of plug-ins. Our CWB is based on the Eclipse [2] Integrated Development Environment (IDE), which is designed as a small kernel containing a plug-in loader for hundreds of plug-ins. The kernel itself is an implementation of a known specification to provide an environment for the plug-ins to execute. This design enables modularity, where discrete chunks of functionality can be reused to build new applications. The minimal set of plug-ins necessary to create a client application is called the Eclipse Rich Client Platform (RCP) [3]; The Eclipse RCP also supports thousands of community-contributed plug-ins, making it a popular development platform for many diverse applications including the Science Activity Planner developed at JPL for the Mars rovers [4] and the scientific experiment tool Gumtree [5]. By leveraging the Eclipse RCP to provide an open, extensible framework, a CWB supports customizations via plug-ins to build rich user applications specific for Earth Science. More importantly, CWB plug-ins can be used by existing science tools built off Eclipse such as IDL or PyDev to provide seamless collaboration functionalities.

Ramachandran, Rahul↗
DOE OSTI2022

Exploring the Use of Novel Spatial Accelerators in Scientific Applications

Driven by the need to find alternative accelerators which can viably replace GPUs in next-generation Supercomputing systems, this paper proposes a methodology to enable agile application/hardware co-design. The application-first methodology provides the ability to come up with design of accelerators while working with real-world workloads, available accelerators, and system software. The iterative design process targets a set of kernels in a workload for performance estimates that can prune the design space for later phases of detailed architectural evaluations. To this effect, in this paper, a novel data-parallel device model is introduced that simulates the latency of performance-sensitive operations in an accelerator including data transfers and kernel computation using multi-core CPUs. The use of off-the-shelf simulators, such as pre-RTL simulator Aladdin or multiple tools available for exploring the design of deep neural network accelerators (e.g., Timeloop) is demonstrated for evaluation of various accelerator designs using applications with realistic inputs. Examples of multiple device configurations that are instantiable in a system are explored to evaluate the performance benefit of deploying novel accelerators. The proposed device is integrated with a programming model and system software to potentially explore the impacts of high-level programming languages/compilers and low-level effects such as task scheduling on multiple accelerators. We analyze our methodology for a set of applications that represent high-performance computing (HPC) and graph analytics. The applications include a computational chemistry kernel realized using tensor contractions, triangle counting, GraphSAGE and Breadth-first Search. These applications include kernels such as dense matrix-dense matrix multiplication, sparse matrix-spare matrix multiplication, and sparse matrix-dense vector multiplication. Our results indicate potential performance benefits and insights for system design by including accelerators that realize these kernels along-side general purpose accelerators.

AI, codesign, Accelerated Computing, Modeling and ↗
DOE OSTI2025

Architecture-Aware Models of AI Engines for High-Performance Matrix Matrix Multiplication

The AI Engine (AIE) architecture, available in systems from mobile SoCs to server-class FPGAs, aims to efficiently execute AI/ML tasks through a two-dimensional array of compute tiles. Previous work on AIEs has explored different approaches to mapping computation across spatial arrays, but the compute kernel running on each tile has not been the focus. Additionally, the AIE-ML architecture introduces memory tiles and omits programmable logic, requiring new approaches to staging and moving data throughout the array. In this work we update analytical models developed for CPUs to produce the design of high performance kernels while introducing new model considerations such as memory structure, throughput, and latency as required by the AIE hardware. We evaluate our models by developing AIE-ML kernels for matrix multiplication in low-precision data types showing performance up to 95% of compute peak for the kernel when data resides in local memory and above 90% of compute peak when data resides in main memory.

Binder, Elliott D. [Carnegie Mellon University, Pi↗
DOE OSTI2020

A High-Throughput Solver for Marginalized Graph Kernels on GPU

Here, we present the design and optimization of a solver for efficient and high-throughput computation of the marginalized graph kernel on General Purpose GPUs. The graph kernel is computed using the conjugate gradient method to solve a generalized Laplacian of the tensor product between a pair of graphs. To cope with the large gap between the instruction throughput and the memory bandwidth of the GPUs, our solver forms the graph tensor product on-the-fly without storing it in memory. This is achieved by using threads in a warp cooperatively to stream the adjacency and edge label matrices of individual graphs by small square matrix blocks called tiles, which are then staged in registers and the shared memory for later reuse. Warps across a thread block can further share tiles via the shared memory to increase data reuse. We exploit the sparsity of the graphs hierarchically by storing only non-empty tiles using a coordinate format and nonzero elements within each tile using bitmaps. We propose a new partition-based reordering algorithm for aggregating nonzero elements of the graphs into fewer but denser tiles to further exploit sparsity. We carry out extensive theoretical analyses on the graph tensor product primitives for tiles of various density and evaluate their performance on synthetic and real-world datasets. Our solver delivers three to four orders of magnitude speedup over existing CPU-based solvers such as GraKeL and GraphKernels. The capability of the solver enables kernel-based learning tasks at unprecedented scales.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

HetArch: Heterogeneous Microarchitectures for Superconducting Quantum Systems

Noisy Intermediate-Scale Quantum Computing (NISQ) has dominated headlines in recent years, with the longer-term vision of Fault-Tolerant Quantum Computation (FTQC) offering significant potential but at currently intractable resource costs and quantum error correction (QEC) overheads. For problems of interest, FTQC will require millions of physical qubits with long coherence times, high-fidelity gates, and compact sizes to surpass classical systems. Just as heterogeneous specialization has offered scaling benefits in classical computing, it is likewise gaining interest in FTQC. However, systematic use of heterogeneity in either hardware or software elements of FTQC systems remains a serious challenge due to the vast design space and the variable physical constraints. This paper meets the challenge of making heterogeneous FTQC design practical by introducing HetArch, a toolbox for designing heterogeneous quantum systems, and using it to explore heterogeneous design scenarios. Using a hierarchical approach, we successively break quantum algorithms into smaller operations (akin to classical application kernels), thus greatly simplifying the design space and resulting tradeoffs. Specializing to superconducting systems, we then design optimized heterogeneous hardware composed of varied superconducting devices, abstracting physical constraints into design rules that enable devices to be assembled into standard cells optimized for specific operations, which, in turn, form heterogeneous modules optimized for quantum subroutines. Finally, we provide a heterogeneous design space exploration framework which reduces the simulation burden by a factor of 10^4 or more and allows us to characterize optimal design points. We use these techniques to design superconducting quantum modules for entanglement distillation, error correction, and code teleportation, reducing error rates by 2.6×, 10.7×, and 3.4× compared to homogeneous systems.

Quantum Computing, Quantum Physics, Computer Archi↗
DOE OSTI2022

Data-driven Minimum Entropy Control for Stochastic Nonlinear Systems using the Cumulant-Generating Function

Here, we present a novel minimum entropy control algorithm for a class of stochastic nonlinear systems subjected to non-Gaussian noises. The entropy control can be considered as an optimization problem for the system randomness attenuation, but the mean value has to be considered separately. To overcome this disadvantage, a new representation of the system stochastic properties was given using the cumulant-generating function based on the moment-generating function, in which the mean value and the entropy was reflected by the shape of the cumulant-generating function. Based on the samples of the system output and control input, a time-variant linear model was identified, and the minimum entropy optimization was transformed to system stabilization. Then, an optimal control strategy was developed to achieve the randomness attenuation, and the boundedness of the controlled system output was analyzed. The effectiveness of the presented control algorithm was demonstrated by a numerical example. In this paper, a data-driven minimum entropy design is presented without pre-knowledge of the system model; entropy optimization is achieved by the system stabilization approach in which the stochastic distribution control and minimum entropy are unified using the same identified structure; and a potential framework is obtained since all the existing system stabilization methods can be adopted to achieve the minimum entropy objective.

42 ENGINEERING↗
DOE OSTI2023

TwoFold: Highly accurate structure and affinity prediction for protein-ligand complexes from sequences

We describe our development of ab initio protein-ligand binding pose prediction models based on transformers and binding affinity prediction models based on the neural tangent kernel (NTK). Folding both protein and ligand, the TwoFold models achieve efficient and quality predictions matching state-of-the-art implementations while additionally reconstructing protein structures. In conclusion, solving NTK models points to a new use case for highly optimized linear solver benchmarking codes on HPC.

60 APPLIED LIFE SCIENCES↗
NASA NTRS1993

Parallel Architectures for Planetary Exploration Requirements (PAPER)

The project's main contributions have been in the area of student support. Throughout the project, at least one, in some cases two, undergraduate students have been supported. By working with the project, these students gained valuable knowledge involving the scientific research project, including the not-so-pleasant reporting requirements to the funding agencies. The other important contribution was towards the establishment of a graduate program in computer science at Hampton University. Primarily, the PAPER project has served as the main research basis in seeking funds from other agencies, such as the National Science Foundation, for establishing a research infrastructure in the department. In technical areas, especially in the first phase, we believe the trip to Jet Propulsion Laboratory, and gathering together all the pertinent information involving experimental computer architectures aimed for planetary explorations was very helpful. Indeed, if this effort is to be revived in the future due to congressional funding for planetary explorations, say an unmanned mission to Mars, our interim report will be an important starting point. In other technical areas, our simulator has pinpointed and highlighted several important performance issues related to the design of operating system kernels for MIMD machines. In particular, the critical issue of how the kernel itself will run in parallel on a multiple-processor system has been addressed through the various ready list organization and access policies. In the area of neural computing, our main contribution was an introductory tutorial package to familiarize the researchers at NASA with this new and promising field zone axes (20). Finally, we have introduced the notion of reversibility in programming systems which may find applications in various areas of space research.

Cezzar, Ruknet↗
NASA NTRS2000

A Reduced Order Model of the Linearized Incompressible Navier-Strokes Equations for the Sensor/Actuator Placement Problem

A reduced order modeling approach of the Navier-Stokes equations is presented for the design of a distributed optimal feedback kernel. This approach is based oil a Krylov subspace method where significant modes of the flow are captured in the model This model is then used in all optimal feedback control design where sensing and actuation is performed oil tile entire flow field. This control design approach yields all optimal feedback kernel which provides insight into the placement of sensors and actuators in the flow field. As all evaluation of this approach, a two-dimensional shear layer and driven cavity flow are investigated.

Allan, Brian G.↗
DOE OSTI2022

ASAP: Automatic Synthesis of Area-Efficient and Precision-Aware CGRAs

Coarse-grained reconfigurable accelerators (CGRAs) are a promising accelerator design choice that strikes a balance between performance and adaptability to different computing patterns across various applications domains. Designing a CGRA for a specific application domain involves enormous software/hardware engineering effort. Recent research works explore loop transformations, functional unit types, network topology, and memory size to identify optimal CGRA designs given a set of kernels from a specific application do- main. Unfortunately, the impact of functional units with different precision support has rarely been investigated. To address this gap, we propose ASAP – a hardware/software co-design framework that automatically identifies and synthesizes optimal precision-aware CGRA for a set of applications of interest. Our evaluation shows that ASAP generates specialized designs 3.2×, 4.21×, and 5.8× more efficient (in terms of performance per unit of energy or area) than non-specialized homogeneous CGRAs, for the scientific computing, embedded, and edge machine learning domains, respectively, with limited accuracy loss. Moreover, ASAP provides more efficient designs than other state-of-the-art synthesis frameworks for specialized CGRAs.

artificial intelligence↗
DOE OSTI2023

Batched Sparse Linear Algebra (Final Report for Subcontract B648960)

This report finalizes design specifications for developing batched kernels for small tensor operations for unassembled matrix-free iterative solvers, batched solvers for partially assembled operators, and batched solvers with support for various sparse formats. The outcome of the project milestones is a set of interfaces to Batched Sparse LA solvers running on hardware accelerators for use in ECP Libraries and Applications. It is part of the development of sparse batched kernels, solvers/preconditioners as well as creating interoperability in xSDK libraries with sparse and dense batched functions to benefit ECP applications. The participants included representatives from ECP libraries (not limited to the xSDK project), applications, and vendors (AMD, Intel, and NVIDIA). Batched sparse linear algebra solvers form the new frontier for algorithmic development and performance engineering. Many applications (ECP and non-ECP alike) require simultaneous solutions of small linear systems of equations that are structurally sparse. To move towards high hardware utilization, it is important to provide these applications with appropriate interfaces to efficient batched sparse solvers running on modern hardware accelerators. We present interface designs in use by HPC software libraries supporting batched sparse linear algebra and the development of sparse batched kernel codes for solvers and preconditioners. We also address the potential interoperability opportunities to keep the software portable between the major hardware accelerators from AMD, Intel, and NVIDIA. The presented interface specifications includes batched band, sparse iterative, and sparse direct solvers. This report summarizes progress in Kokkos Kernels and the xSDK libraries MAGMA, Ginkgo, hypre, SUNDIALS, and SuperLU_dist.

97 MATHEMATICS AND COMPUTING↗