Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “heterogeneous 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 289 records · Page 16
All sourcesNASA NTRSDOE OSTI
NASA NTRS2003

Mesoscale Computational Investigation of Shocked Heterogeneous Materials with Application to Large Impact Craters

The propagation of shock waves through target materials is strongly influenced by the presence of small-scale structure, fractures, physical and chemical heterogeneities. Pre-existing fractures often create craters that appear square in outline (e.g. Meteor Crater). Reverberations behind the shock from the presence of physical heterogeneity have been proposed as a mechanism for transient weakening of target materials. Pre-existing fractures can also affect melt generation. In this study, we are attempting to bridge the gap in numerical modeling between the micro-scale and the continuum, the so-called meso-scale. To accomplish this, we are developing a methodology to be used in the shock physics hydrocode (CTH) using Monte-Carlo-type methods to investigate the shock properties of heterogeneous materials. By comparing the results of numerical experiments at the micro-scale with experimental results and by using statistical techniques to evaluate the performance of simple constitutive models, we hope to embed the effect of physical heterogeneity into the field variables (pressure, stress, density, velocity) allowing us to directly imprint the effects of micro-scale heterogeneity at the continuum level without incurring high computational cost.

Crawford, D. A.↗
NASA NTRS1994

Accommodating Heterogeneity in a Debugger for Distributed Computations

In an ongoing project at NASA Ames Research Center, we are building debugger for distributed computations running on a heterogeneous set of machines. Historically, such debuggers have been built as front-ends to existing source-level debuggers on the target platforms. In effect, these back-end debuggers are providing a collection of debugger services to a client. The major drawback is that because of inconsistencies among the back-end debuggers, the front-end must use a different protocol when talking to each back-end debugger. This can make the front-end quite complex. We have avoided this complexity problem by defining the client-server debugger protocol. While it does require vendors to adapt their existing debugger code to meet the protocol, vendors are generally interested in doing so because the approach has several advantages. In addition to solving the heterogenous platform debugging problem, it will be pos~sible to write interesting debugger user interfaces that can be easily ported across a variety of machines. This will likely encourage investment in application-domain specific debuggers. In fact, the user interface of our debugger will be geared to scientists developing computational fluid dynamics codes. This paper describes some of the problems encountered in developing a portable debugger for heterogenous, distributed computing and how the architecture of our debugger avoids them. It then provides a detailed description of the debugger client-server protocol. Some of the more interesting attributes of the protocol are: (1) It is object-oriented; (2) It uses callback functions to capture the asynchronous nature of debugging in a procedural fashion; (3) It contains abstractions, such as in-line instrumentation, for the debugging of computationally intensive programs; (4) For remote debugging, it has operations that enable the implementor to optimize message passing traffic between client and server. The soundness of the protocol is being tested through the implementation of a prototype client and server pair.

Hood, Robert↗
NASA NTRS1996

Effect of Pt Doping on Nucleation and Crystallization in Li2O.2SiO2 Glass: Experimental Measurements and Computer Modeling

Heterogeneous nucleation and its effects on the crystallization of lithium disilicate glass containing small amounts of Pt are investigated. Measurements of the nucleation frequencies and induction times with and without Pt are shown to be consistent with predictions based on the classical nucleation theory. A realistic computer model for the transformation is presented. Computed differential thermal analysis data (such as crystallization rates as a function of time and temperature) are shown to be in good agreement with experimental results. This modeling provides a new, more quantitative method for analyzing calorimetric data.

Narayan, K. Lakshmi↗
DOE OSTI2021

Extending C++ for Heterogeneous Quantum-Classical Computing

In this report we present qcor - a language extension to C++ and compiler implementation that enables heterogeneous quantum-classical programming, compilation, and execution in a single-source context. Our work provides a first-of-its-kind C++ compiler enabling high-level quantum kernel (function) expression in a quantum-language agnostic manner, as well as a hardware-agnostic, retargetable compiler workflow targeting a number of physical and virtual quantum computing backends. qcor leverages novel Clang plugin interfaces and builds upon the XACC system-level quantum programming framework to provide a state-of-the-art integration mechanism for quantum-classical compilation that leverages the best from the community at-large. qcor translates quantum kernels ultimately to the XACC intermediate representation, and provides user-extensible hooks for quantum compilation routines like circuit optimization, analysis, and placement. This work details the overall architecture and compiler workflow for qcor, and provides a number of illuminating programming examples demonstrating its utility for near-term variational tasks, quantum algorithm expression, and feed-forward error correction schemes.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

diffReplication - An Energy-Aware Fault Tolerance Model for Silent Error Detection and Mitigation in Heterogeneous Extreme-scale Computing Environment

At extreme scale, the frequency of silent errors – a class of errors that remain undetected by low-level error detection mechanisms – increases significantly with the computational complexity of the application and the scale of the computing infrastructure. As hardware and software advances are made to usher in the next scientific era of computing, developing new approaches to mitigate the impact of silent errors remains a challenging problem. In this work, we propose an energy-aware fault-tolerance model, referred to diffReplication to overcome silent errors. In the proposed model, the main process is associated with one replica that executes at the same rate as the main process, and one diffReplica that is executed at a fraction of the main process' execution rate. If the main and its replica reach consensus at the end of a computation phase, the state of the diffReplica is updated and computation is resumed. If the synchronization attempt results in a disagreement, however, the diffReplica increases its execution speed to complete the computation and quickly reach the synchronization barrier. Assuming a single error over any given synchronization interval, a majority voting is used to reach consensus and tolerate silent errors. To further enhance its performance, diffReplication is augmented with speculative execution, whereby the main or its fast replica is selected to continue execution without waiting for the diffReplica. The selection process is based on the previous behaviour of the main and its replica. A performance analysis study is carried out to assess the performance of diffReplication, in terms of the energy saving and time-to-completion reduction achieved by the diffReplication scheme. The experiment shows that speculative execution reduces the time to completion with additional energy, and dynamic decision-making balances the energy consumption and time to completion.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Characterizing the performance of node-aware strategies for irregular point-to-point communication on heterogeneous architectures

Supercomputer architectures are trending toward higher computational throughput due to the inclusion of heterogeneous compute nodes. These multi-GPU nodes increase on-node computational efficiency, while also increasing the amount of data to be communicated and the number of potential data flow paths. In this work, we characterize the performance of irregular point-to-point communication with MPI on heterogeneous compute environments through performance modeling, demonstrating the limitations of standard communication strategies for both device-aware and staging-through-host communication techniques. Presented models suggest staging communicated data through host processes then using node-aware communication strategies for high inter-node message counts. Notably, the models also predict that node-aware communication utilizing all available CPU cores to communicate inter-node data leads to the most performant strategy when communicating with a high number of nodes. Furthermore, model validation is provided via a case study of irregular point-to-point communication patterns in distributed sparse matrix–vector products. Importantly, we include a discussion on the implications model predictions have on communication strategy design for emerging supercomputer architectures.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

Heterogeneous CPU+GPU-Enabled Simulations for DFTB Molecular Dynamics of Large Chemical and Biological Systems

We introduce a new heterogeneous CPU+GPU-enhanced DFTB approach for the routine and efficient simulation of large chemical and biological systems. Here, compared to homogenous computing with conventional CPUs, heterogeneous computing approaches exhibit substantial performance with only a modest increase in power consumption, both of which are essential to upcoming exascale computing initiatives. We show that DFTB-based molecular dynamics is a natural candidate for heterogeneous computing since the computational bottleneck in these simulations is the diagonalization of the Hamiltonian matrix, which is performed several times during a single molecular dynamics trajectory. To thoroughly test and understand the performance of our heterogeneous CPU+GPU approach, we examine a variety of algorithmic implementations, benchmarks of different hardware configurations, and applications of this methodology on several large chemical and biological systems. Finally, to demonstrate the capability of our implementation, we conclude with a large-scale DFTB MD simulation of explicitly solvated HIV protease (3,974 atoms total) as a proof-of-concept example of an extremely large/complex system which, to the best of our knowledge, is the first time that an entire explicitly-solvated protein has been treated at a quantum-based MD level of detail.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2025

CAFE AU LAIT: Compute-Aware Federated Augmented Low-Rank AI Training

Federated finetuning is crucial for unlocking the knowledge embedded in pretrained Large Language Models (LLMs) when data are geographically distributed across clients. Unlike finetuning with data from a single institution, federated finetuning allows collaboration across multiple institutions, enabling the utilization of diverse and decentralized datasets while preserving data privacy. Given the high computing costs of LLM training and the emphasis on energy efficiency in Federated Learning (FL), Low-Rank Adaptation (LoRA) has emerged as a widely adopted algorithm due to its significantly reduced number of trainable parameters. However, this assumes that all data silos have the necessary computing resources to compute local updates of LLMs. Nevertheless, in practice, the computing resources across clients are highly heterogeneous: while some may have access to hundreds of GPUs, others might have limited or no GPU access. Recently, federated finetuning using synthetic data has been proposed, allowing clients to participate in a collaborative training run without training LLMs locally. However, our experimental results reveal a performance gap between models trained using synthetic data and those trained using local updates. Motivated by the observed heterogeneity in computing resources and the performance gap, we propose a novel two-stage algorithm that leverages the storage and computing capabilities of a strong server. In the first stage, under the coordination of the strong server, clients with limited computing resources collaborate to generate synthetic data, which is transferred to and stored on the strong server. In the second stage, the strong server uses this synthetic data on behalf of the resource-constrained clients to perform federated LoRA finetuning alongside clients with sufficient computing resources. This approach ensures that all clients can participate in the finetuning process. Experimental results demonstrate that incorporating local updates from even a small fraction of clients improves performance compared to using synthetic data for all clients. Furthermore, we incorporate the Gaussian mechanism in both stages to guarantee client-level differential privacy.

Wang, Jiayi [ORNL]↗
DOE OSTI2019

SYCL for Monte Carlo Transport [Slides]

SYCL is a cross-platform abstraction layer that builds on the underlying concepts, portability and efficiency of OpenCL that enables code for heterogeneous processors to be written in a “single-source” style using completely standard C++.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2025

Flexible and Effective Object Tiering for Heterogeneous Memory Systems

Computing platforms that package multiple types of memory, each with their own performance characteristics, are quickly becoming mainstream. To operate efficiently, heterogeneous memory architectures require new data management solutions that are able to match the needs of each application with an appropriate type of memory. As the primary generators of memory usage, applications create a great deal of information that can be useful for guiding memory management, but the community still lacks tools to collect, organize, and leverage this information effectively. To address this gap, this work introduces a novel software framework that collects and analyzes object-level information to guide memory tiering. The framework includes tools to monitor the capacity and usage of individual data objects, routines that aggregate and convert this information into tier recommendations for the host platform, and mechanisms to enforce these recommendations according to user-selected policies. Moreover, the developed tools and techniques are fully automatic, work on standard Linux systems, and do not require modification or recompilation of existing software. Using this framework, this study evaluates and compares the impact of a variety of design choices for memory tiering, including different policies for prioritizing objects for the fast memory tier as well as the frequency and timing of migration events. In conclusion, the results, collected on a modern Intel platform with conventional DDR4 SDRAM as well as Intel Optane NVRAM, show that guiding data tiering with object-level information can enable significant performance and efficiency benefits compared with standard hardware- and software-directed data-tiering strategies for a diverse set of memory-intensive workloads.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

Parallelization of a Proxy Transport App Using ComputeCPP and SYCL

This document details the attempts to parallelize a multi-group Monte Carlo proxy app using SYCL and the ComputeCPP Community Edition compiler. SYCL is a standard set by the Khronos Group and is described as a, “cross-platform abstraction layer that builds on the underlying concepts, portability and efficiency of OpenCL that enables code for heterogeneous processors to be written in a “single-source” style using completely standard C++.” ComputeCpp Community Edition (CE) is a heterogeneous parallel programming platform that provides a conformant implementation of the SYCL™ 1.2.1 Khronos specification. ComputeCPP provides a compiler and an SDK containing a build system to improve the usability of the compiler. The supported OpenCL 1.2 platforms for ComputeCpp are AMD® and Intel®, with experimental support for PTX which is used on Nvidia graphics cards as well as experimental support for ARM chips. All testing reported in this document was done on the Darwin computer cluster using an Nvidia GTX Titan X graphics card. Since ComputeCPP does not support using this card with OpenCL, ComputeCPP’s experimental PTX64 bit-code generation was used to generate device code.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2022

Bridging the Gap between the X-ray Absorption Spectroscopy and the Computational Catalysis Communities in Heterogeneous Catalysis: A Perspective on the Current and Future Research Directions

X-ray absorption spectroscopy (XAS) [extended X-ray absorption fine structure (EXAFS) and X-ray absorption near-edge structure (XANES)] is a key technique within the heterogeneous catalysis community to probe the structure and properties of the active site(s) for a diverse range of catalytic materials. Furthermore, the interpretation of the raw experimental data to derive an atomistic picture of the catalyst requires modeling and analysis; the EXAFS data are compared to a model, and a goodness of fit parameter is used to judge the best fit. This EXAFS modeling can often be nontrivial and time-consuming; overcoming or improving these limitations remains a central challenge for the community. Considering these limitations, this Perspective highlights how recent developments in analysis software, increased availability of reliable computational models, and application of data science tools can be used to improve the speed, accuracy, and reliability of EXAFS interpretation. In particular, we emphasize the advantages of combining theory and EXAFS as a unified technique that should be treated as a standard (when applicable) to identify catalytic sites and not two separate complementary methods. Building on the recent trends in the computational catalysis community, we also present a community-driven approach to adopt FAIR Guiding Principles for the collection, analysis, dissemination, and storage of XAS data. Written with both the experimental and theory audience in mind, we provide a unified roadmap to foster collaborations between the two communities.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2016

Multiscale analysis and computation for flows in heterogeneous media

Our work in this project is aimed at making fundamental advances in multiscale methods for flow and transport in highly heterogeneous porous media. The main thrust of this research is to develop a systematic multiscale analysis and efficient coarse-scale models that can capture global effects and extend existing multiscale approaches to problems with additional physics and uncertainties. A key emphasis is on problems without an apparent scale separation. Multiscale solution methods are currently under active investigation for the simulation of subsurface flow in heterogeneous formations. These procedures capture the effects of fine-scale permeability variations through the calculation of specialized coarse-scale basis functions. Most of the multiscale techniques presented to date employ localization approximations in the calculation of these basis functions. For some highly correlated (e.g., channelized) formations, however, global effects are important and these may need to be incorporated into the multiscale basis functions. Other challenging issues facing multiscale simulations are the extension of existing multiscale techniques to problems with additional physics, such as compressibility, capillary effects, etc. In our project, we explore the improvement of multiscale methods through the incorporation of additional (single-phase flow) information and the development of a general multiscale framework for flows in the presence of uncertainties, compressible flow and heterogeneous transport, and geomechanics. We have considered (1) adaptive local-global multiscale methods, (2) multiscale methods for the transport equation, (3) operator-based multiscale methods and solvers, (4) multiscale methods in the presence of uncertainties and applications, (5) multiscale finite element methods for high contrast porous media and their generalizations, and (6) multiscale methods for geomechanics. Below, we present a brief overview of each of these contributions.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Optimizing High-Throughput Inference on Graph Neural Networks at Shared Computing Facilities with the NVIDIA Triton Inference Server

Abstract With machine learning applications now spanning a variety of computational tasks, multi-user shared computing facilities are devoting a rapidly increasing proportion of their resources to such algorithms. Graph neural networks (GNNs), for example, have provided astounding improvements in extracting complex signatures from data and are now widely used in a variety of applications, such as particle jet classification in high energy physics (HEP). However, GNNs also come with an enormous computational penalty that requires the use of GPUs to maintain reasonable throughput. At shared computing facilities, such as those used by physicists at Fermi National Accelerator Laboratory (Fermilab), methodical resource allocation and high throughput at the many-user scale are key to ensuring that resources are being used as efficiently as possible. These facilities, however, primarily provide CPU-only nodes, which proves detrimental to time-to-insight and computational throughput for workflows that include machine learning inference. In this work, we describe how a shared computing facility can use the NVIDIA Triton Inference Server to optimize its resource allocation and computing structure, recovering high throughput while scaling out to multiple users by massively parallelizing their machine learning inference. To demonstrate the effectiveness of this system in a realistic multi-user environment, we use the Fermilab Elastic Analysis Facility augmented with the Triton Inference Server to provide scalable and high-throughput access to a HEP-specific GNN and report on the outcome.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
NASA NTRS2008

Transformation of OODT CAS to Perform Larger Tasks

A computer program denoted OODT CAS has been transformed to enable performance of larger tasks that involve greatly increased data volumes and increasingly intensive processing of data on heterogeneous, geographically dispersed computers. Prior to the transformation, OODT CAS (also alternatively denoted, simply, 'CAS') [wherein 'OODT' signifies 'Object-Oriented Data Technology' and 'CAS' signifies 'Catalog and Archive Service'] was a proven software component used to manage scientific data from spaceflight missions. In the transformation, CAS was split into two separate components representing its canonical capabilities: file management and workflow management. In addition, CAS was augmented by addition of a resource-management component. This third component enables CAS to manage heterogeneous computing by use of diverse resources, including high-performance clusters of computers, commodity computing hardware, and grid computing infrastructures. CAS is now more easily maintainable, evolvable, and reusable. These components can be used separately or, taking advantage of synergies, can be used together. Other elements of the transformation included addition of a separate Web presentation layer that supports distribution of data products via Really Simple Syndication (RSS) feeds, and provision for full Resource Description Framework (RDF) exports of metadata.

Mattmann, Chris↗