Results for “parallel performance” — Page 17

DOE OSTI2023

PhytoOracle: Scalable, modular phenomics data processing pipelines

As phenomics data volume and dimensionality increase due to advancements in sensor technology, there is an urgent need to develop and implement scalable data processing pipelines. Current phenomics data processing pipelines lack modularity, extensibility, and processing distribution across sensor modalities and phenotyping platforms. To address these challenges, we developed PhytoOracle (PO), a suite of modular, scalable pipelines for processing large volumes of field phenomics RGB, thermal, PSII chlorophyll fluorescence 2D images, and 3D point clouds. PhytoOracle aims to ( i ) improve data processing efficiency; ( ii ) provide an extensible, reproducible computing framework; and ( iii ) enable data fusion of multi-modal phenomics data. PhytoOracle integrates open-source distributed computing frameworks for parallel processing on high-performance computing, cloud, and local computing environments. Each pipeline component is available as a standalone container, providing transferability, extensibility, and reproducibility. The PO pipeline extracts and associates individual plant traits across sensor modalities and collection time points, representing a unique multi-system approach to addressing the genotype-phenotype gap. To date, PO supports lettuce and sorghum phenotypic trait extraction, with a goal of widening the range of supported species in the future. At the maximum number of cores tested in this study (1,024 cores), PO processing times were: 235 minutes for 9,270 RGB images (140.7 GB), 235 minutes for 9,270 thermal images (5.4 GB), and 13 minutes for 39,678 PSII images (86.2 GB). These processing times represent end-to-end processing, from raw data to fully processed numerical phenotypic trait data. Repeatability values of 0.39-0.95 (bounding area), 0.81-0.95 (axis-aligned bounding volume), 0.79-0.94 (oriented bounding volume), 0.83-0.95 (plant height), and 0.81-0.95 (number of points) were observed in Field Scanalyzer data. We also show the ability of PO to process drone data with a repeatability of 0.55-0.95 (bounding area).

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2020

NMDQi Nuclear Materials Discovery and Qualification Initiative Conference Overview

The Nuclear Materials Discovery and Qualification Initiative (NMDQi) is designed to accelerate nuclear materials qualification to fulfill the promises of early and advanced reactor technologies as a safe, clean, and low-cost baseload energy. Materials development and qualification in the nuclear industry is by definition challenging due to stringent safety requirements, limited availability of specialized facilities for materials irradiation and testing, and the challenging high-temperature, high-radiation environment. NMDQi will establish tools and capabilities that will greatly accelerate the nuclear fuels and materials development process. These tools will provide both computationally informed insights and high-throughput infrastructure with the goal of completing qualification in a single pass. In this approach, materials must be fabricated with a range of properties of interest so that materials performance can be examined in parallel, rather than with multiple discrete specimens. Advanced manufacturing (AM) techniques, which can produce complex component geometries, microstructures, and compositions, are ideal. The improved process monitoring and control that AM can provide are ideal for ensuring and reducing material variability, which will be particularly important as standards committees pursue regulations for AM components. However, gaining these benefits requires overcoming the substantial barrier of qualifying AM processes and the resulting materials and components for supporting research campaigns and eventually nuclear service.

36 MATERIALS SCIENCE↗

DOE OSTI2020

Orchestrating Fault Prediction with Live Migration and Checkpointing

Checkpoint/Restart (C/R) is widely used to provide fault tolerance on High-Performance Computing (HPC) systems. However, Parallel File System (PFS) overhead and failure uncertainty cause significant application overhead. This paper develops an adaptive multi-level C/R model that incorporates a failure prediction and analysis model, which orchestrates failure prediction, checkpointing, checkpoint frequency, and proactive live migration along with the additional benefit of Burst Buffers (BB). It effectively reduces the overheads due to failures, checkpointing, and recovery. Simulation results for the Summit supercomputer yield a reduction of ~20%-86% in application overhead due to BBs, orchestrated failure prediction, and migration. We also observe a ~29% decrease in checkpoint writes to BBs, which can increase the longevity of the BB storage devices.

Behera, Subhendu↗

DOE OSTI2021

High-Level Synthesis of Parallel Specifications Coupling Static and Dynamic Controllers

The increased need for efficient ways to implement domain-specific accelerators is driving design methodologies towards the use of abstractions higher than the Register Transfer Level (RTL). In this scenario, High Level Synthesis (HLS) plays a significant role by enabling the automatic generation of custom hardware accelerators starting from high level descriptions (e.g., C code). Conventional HLS tools exploit parallelism mostly at the Instruction Level (ILP). They statically schedule the input specifications, and build centralized Finite State Machine (FSM) controllers. However, aggressive exploitation of ILP in many applications has diminishing returns and, usually, centralized approaches do not efficiently exploit coarser parallelism because FSMs are inherently serial. In this paper we present a HLS framework able to synthesize applications that, beside ILP, also expose Task Level Parallelism (TLP). An application can expose TLP through annotations that identify the parallel functions (i.e., tasks). To generate accelerators that efficiently execute concur- rent tasks, we need to solve several issues: devise a mechanism to support concurrent execution flows, exploit memory parallelism, and manage synchronization. To support concurrent execution flows, we introduce a novel adaptive controller. The adaptive controller is composed of a set of interacting control elements that independently manage the execution of a single operation or function call. These control elements check dependencies and resource constraints at runtime, enabling as soon as possible execution. To support parallel access to shared memories and synchronization, we introduce a novel Hierarchical Memory Interface (HMI). With respect to previous solutions, the proposed interface supports multi-ported memories and atomic memory operations, which commonly occur in parallel programming. Our framework can generate the hardware implementation of C functions by employing two different approaches, depending on its characteristics. If a function exposes TLP, then the framework generates hardware implementations based on the adaptive controller. Otherwise, the framework implements the function by exploiting a more conventional FSM approach, which is optimized for ILP exploitation. We evaluate our framework on a set of parallel applications, and show substantial performance improvements (average speedup of 4.7) with limited area over- heads (average area increase of 5.48 times).

Castellana, Vito G.↗

DOE OSTI2021

High-Level Synthesis of Parallel Specifications Coupling Static and Dynamic Controllers

The increased need for efficient ways to implement domain-specific accelerators is driving design methodologies towards the use of abstractions higher than the Register Transfer Level (RTL). In this scenario, High Level Synthesis (HLS) plays a significant role by enabling the automatic generation of custom hardware accelerators starting from high level descriptions (e.g., C code). Conventional HLS tools exploit parallelism mostly at the Instruction Level (ILP). They statically schedule the input specifications, and build centralized Finite State Machine (FSM) controllers. However, aggressive exploitation of ILP in many applications has diminishing returns and, usually, centralized approaches do not efficiently exploit coarser parallelism because FSMs are inherently serial. In this paper we present a HLS framework able to synthesize applications that, beside ILP, also expose Task Level Parallelism (TLP). An application can expose TLP through annotations that identify the parallel functions (i.e., tasks). To generate accelerators that efficiently execute concur- rent tasks, we need to solve several issues: devise a mechanism to support concurrent execution flows, exploit memory parallelism, and manage synchronization. To support concurrent execution flows, we introduce a novel adaptive controller. The adaptive controller is composed of a set of interacting control elements that independently manage the execution of a single operation or function call. These control elements check dependencies and resource constraints at runtime, enabling as soon as possible execution. To support parallel access to shared memories and synchronization, we introduce a novel Hierarchical Memory Interface (HMI). With respect to previous solutions, the proposed interface supports multi-ported memories and atomic memory operations, which commonly occur in parallel programming. Our framework can generate the hardware implementation of C functions by employing two different approaches, depending on its characteristics. If a function exposes TLP, then the framework generates hardware implementations based on the adaptive controller. Otherwise, the framework implements the function by exploiting a more conventional FSM approach, which is optimized for ILP exploitation. We evaluate our framework on a set of parallel applications, and show substantial performance improvements (average speedup of 4.7) with limited area over- heads (average area increase of 5.48 times).

Castellana, Vito G.↗

DOE OSTI2023

Neutron Absorber Plate Characterization Plan for Criticality Experiments Design

After being used in nuclear installations, depleted fuel can still be highly reactive and must be handled securely to prevent any radiological or criticality concerns. In particular, spent fuel from use in nuclear power reactors must be stored and transported in specifically designed containers using neutron absorber materials to prevent criticality. Various neutron absorber material types exist and are manufactured by various entities, as thoroughly described in the Handbook of Neutron Absorber Materials for Spent Nuclear Fuel Storage and Transportation Applications written by EPRI. Presently, one of the most modern and most widely used types of neutron absorber material contains particles of boron carbide, or B 4 C, embedded in aluminum matrix: Boralcan, manufactured by Rio Tinto. It is very important for the community to know as much as possible about such neutron absorber materials. Therefore, in the recent years, a US Department of Energy National Nuclear Security Administration–Nuclear Criticality Safety Program funded project initiated design of an experiment that places Boralcan neutron-absorbing plates in an established critical assembly using low-enriched uranium fuel at the Sandia Pulsed Reactor Facility/Critical Experiments (SPRF/CX) apparatus at Sandia National Laboratories. The goal of the experiment is to produce high-quality benchmark data to submit to the International Criticality Safety Benchmark Evaluation Project (ICSBEP), for use in validating calculational tools and nuclear data by criticality safety analysts. The project, named IER-554, is currently in its final design stage, following a successful preliminary design. In the work documented in the design study, ten critical configurations using Boralcan neutron absorber plates were designed, and the experiment was proven to be feasible, with a predicted low k eff uncertainty around 100 pcm. An overview of the modeled cutout of the critical assembly with a Boralcan plate is shown in Figure 1, representing one of the configurations planned for the critical experiments. Before the plates are inserted in the critical assembly, it is necessary to know more about their composition and uniformity. This summary focuses on the plate characterization plans. Each plate will undergo (1) neutron transmission measurements at different locations to determine the 10 B areal density and (2) an in-depth x-ray computed tomography (XCT) examination to obtain the exact Sizes and distribution of the B4C powder particles inside the plates. In parallel, plate modeling studies are performed with a goal to determine the validity of the currently used approximation of modeling the neutron absorber plates as a homogeneous mixture of Aluminum 1100 alloy and B4C— instead of explicitly modeling the B4C particles. By using the experimental 10 B areal density measurements, and the exact size and location of the B4C particles obtained by XCT, a plate model can theoretically be built that reproduces the plate with extremely high fidelity. The results of this modeling study could increase the confidence of the criticality safety community in its modeling methods when using this type of neutron absorber material, and the industry could use these validations to change the boron loading credit limits from the U.S. Nuclear Regulatory Commission standard review plan for dry cask storage of spent nuclear fuel. The modeling calculations are performed with SCALE 6.3.0 using the KENO V.a sequence for criticality calculations with the ENDF/B-VIII.0 continuous-energy cross section library.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

DOE OSTI2024

Reductive Analysis with Compiler-Guided Large Language Models for Input-Centric Code Optimizations

Input-centric program optimization aims to optimize code by considering the relations between program inputs and program behaviors. Despite its promise, a long-standing barrier for its adoption is the difficulty of automatically identifying critical features of complex inputs. This paper introduces a novel technique, reductive analysis through compiler-guided Large Language Models (LLMs), to solve the problem through a synergy between compilers and LLMs. It uses a reductive approach to overcome the scalability and other limitations of LLMs in program code analysis. The solution, for the first time, automates the identification of critical input features without heavy instrumentation or profiling, cutting the time needed for input identification by 44× (or 450× for local LLMs), reduced from 9.6 hours to 13 minutes (with remote LLMs) or 77 seconds (with local LLMs) on average, making input characterization possible to be integrated into the workflow of program compilations. Optimizations on those identified input features show similar or even better results than those identified by previous profiling-based methods, leading to optimizations that yield 92.6% accuracy in selecting the appropriate adaptive OpenMP parallelization decisions, and 20-30% performance improvement of serverless computing while reducing resource usage by 50-60%.

Input-Centric Optimization↗

DOE OSTI2023

Failure analysis of nuclear transient-tested UN tristructural isotropic fuel particles in a 3D printed SiC matrix

Fully ceramic microencapsulated fuel elements containing UN tristructural isotropic (TRISO) fuel particles within a 3D printed SiC matrix were subjected to transient testing with varying energy depositions. Detailed post-irradiation examinations were performed, including leaching in hot HNO3 and post-leaching X-ray computed tomography, to quantify the percentage of failed TRISO particles and crack propagation within the particles and surrounding fuel matrix. In parallel, detailed finite element analyses were performed for comparison with experimental findings and to better evaluate transient failure modes. The lowest transient energy deposition—which still exceeded bounding values for high-temperature gas-cooled reactor applications—resulted in no detectable TRISO particle failures or matrix cracking, which was consistent with the simulations. Simulations of the higher-energy transients for which significant TRISO particle failure was expected were generally able to reproduce the transient temperatures and matrix cracking. Thus, the TRISO particle failures were explained based on the effects of local SiC matrix thickness and porosity. Results generally confirmed the high strength of the additively manufactured SiC matrix but also affirmed the need for a modified UN TRISO architecture to prevent SiC matrix cracks from propagating through TRISO layers. This unique failure mode has not historically been considered for TRISO fuels contained in weaker graphite matrices.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

DOE OSTI2024

Pretraining Billion-Scale Geospatial Foundational Models on Frontier

As AI workloads increase in scope, generalization capability becomes challenging for small task-specific models and their demand for large amounts of labeled training samples increases. On the contrary, Foundation Models (FMs) are trained with internet-scale unlabeled data via self-supervised learning and have been shown to adapt to various tasks with minimal fine-tuning. Although large FMs have demonstrated significant impact in natural language processing and computer vision, efforts toward FMs for geospatial applications have been restricted to smaller size models, as pretraining larger models requires very large computing resources equipped with state-of-the-art hardware accelerators. Current satellite constellations collect 100+TBs of data a day, resulting in images that are billions of pixels and multimodal in nature. Such geospatial data poses unique challenges opening up new opportunities to develop FMs. We investigate billion scale FMs and HPC training profiles for geospatial applications by pretraining on publicly available data. We studied from end-to-end the performance and impact in the solution by scaling the model size. Our larger 3B parameter size model achieves up to 30% improvement in top1 scene classification accuracy when comparing a 100M parameter model. Moreover, we detail performance experiments on the Frontier supercomputer, America's first exascale system, where we study different model and data parallel approaches using PyTorch's Fully Sharded Data Parallel library. Specifically, we study variants of the Vision Transformer architecture (ViT), conducting performance analysis for ViT models with size up to 15B parameters. By discussing throughput and performance bottlenecks under different parallelism configurations, we offer insights on how to leverage such leadership-class HPC resources when developing large models for geospatial imagery applications.

Tsaris, Aristeidis (aris)↗

DOE OSTI2021

Scaling theory of three-dimensional magnetic reconnection spreading

We develop a first-principles scaling theory of the spreading of three-dimensional (3D) magnetic reconnection of finite extent in the out of plane direction. This theory addresses systems with or without an out of plane (guide) magnetic field, and with or without Hall physics. The theory reproduces known spreading speeds and directions with and without guide fields, unifying previous knowledge in a single theory. New results include: (1) Reconnection spreads in a particular direction if an x-line is induced at the interface between reconnecting and non-reconnecting regions, which is controlled by the out of plane gradient of the electric field in the outflow direction. (2) The spreading mechanism for anti-parallel collisionless reconnection is convection, as is known, but for guide field reconnection it is magnetic field bending. We confirm the theory using 3D two-fluid and resistive-magnetohydrodynamics simulations. (3) The theory explains why anti-parallel reconnection in resistive-magnetohydrodynamics does not spread. (4) The simulation domain aspect ratio, associated with the free magnetic energy, influences whether reconnection spreads or convects with a fixed x-line length. (5) We perform a simulation initiating anti-parallel collisionless reconnection with a pressure pulse instead of a magnetic perturbation, finding spreading is unchanged rather than spreading at the magnetosonic speed as previously suggested. The results provide a theoretical framework for understanding spreading beyond systems studied here, and are important for applications including two-ribbon solar flares and reconnection in Earth’s magnetosphere.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

DOE OSTI2024

Parallel Implementation of the Density Matrix Renormalization Group Method Achieving a Quarter petaFLOPS Performance on a Single DGX-H100 GPU Node

Not Available

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2020

Xyce Case Study

With the elimination of underground nuclear testing and declining defense budgets, science-based stockpile stewardship requires increased reliance on high performance modeling and simulation of weapon systems. Today's weapon systems are comprised of various electrical components and systems. As a result, there is a need for tools that will allow the use of massively parallel modeling and simulation techniques on high performance computers in existing and future weapons' electrical systems models. The Xyce Parallel Electronic Simulator is a SPICE (Simulation Program with Integrated Circuit Emphasis)- compatible circuit simulator designed to run on large-scale parallel computing platforms, though it can also execute efficiently on a variety of architectures including single processor workstations. As a mature platform for large-scale parallel circuit simulation, Xyce supports standard capabilities available in commercial simulators, in addition to various devices and models specific to Sandia's needs. Specifically, Xyce aids in the design and verification of electrical and electronic circuits and systems prior to weapons' manufacturing and deployment.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗

DOE OSTI2020

Xyce Case Study

With the elimination of underground nuclear testing and declining defense budgets, science-based stockpile stewardship requires increased reliance on high performance modeling and simulation of weapon systems. Today's weapon systems are comprised of various electrical components and systems. As a result, there is a need for tools that will allow the use of massively parallel modeling and simulation techniques on high performance computers in existing and future weapons' electrical systems models. The Xyce Parallel Electronic Simulator is a SPICE (Simulation Program with Integrated Circuit Emphasis)- compatible circuit simulator designed to run on large-scale parallel computing platforms, though it can also execute efficiently on a variety of architectures including single processor workstations. As a mature platform for large-scale parallel circuit simulation, Xyce supports standard capabilities available in commercial simulators, in addition to various devices and models specific to Sandia's needs. Specifically, Xyce aids in the design and verification of electrical and electronic circuits and systems prior to weapons' manufacturing and deployment.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗

DOE OSTI2021

Impact of Blockchain Delay on Grid-Tied Solar Inverter Performance

This paper investigates the impact of the delay resulting from a blockchain, a promising security measure, for a hierarchical control system of inverters connected to the grid. The blockchain communication network is designed at the secondary control layer for resilience against cyberattacks. To represent the latency in the communication channel, a model is developed based on the complexity of the blockchain framework. Taking this model into account, this work evaluates the plant’s performance subject to communication delays, introduced by the blockchain, among the hierarchical control agents. In addition, this article considers an optimal model-based control strategy that performs the system’s internal control loop. The work shows that the blockchain’s delay size influences the convergence of the power supplied by the inverter to the reference at the point of common coupling. In the results section, real-time simulations on OPAL-RT are performed to test the resilience of two parallel inverters with increasing blockchain complexity.

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

DOE OSTI2025

Toward a persistent event-streaming system for high-performance computing applications

High-performance computing (HPC) applications have traditionally relied on parallel file systems and file transfer services to manage data movement and storage. Alternative approaches have been proposed that use direct communications between application components, trading persistence and fault tolerance for speed. Event-driven architectures, as popularized in enterprise contexts, present a compelling middle ground, avoiding the performance cost and API constraints of parallel file systems while retaining persistence and offering impedance matching between application components. However, adapting streaming frameworks to HPC workloads requires addressing challenges unique to HPC systems. This paper investigates the potential for a streaming framework designed for HPC infrastructures and use cases. We introduce Mofka, a persistent event-streaming framework designed specifically for HPC environments. Mofka combines the capabilities of a traditional streaming service with optimizations tailored to the HPC context, such as support for massively multicore nodes, efficient scaling for large producer-consumer workflows, RDMA-enabled high-performance network communications, specialized network fabrics with multiple links per node, and efficient handling of large scientific data payloads. Built using the Mochi suite of HPC data service components, Mofka provides a lightweight, modular, and high-performance solution for persistent streaming in HPC systems. We present the architecture of Mofka and evaluate its performance against Kafka and Redpanda using benchmarks on diverse platforms, including Argonne's Polaris and Oak Ridge's Frontier supercomputers, showing up to 8× improvement in throughput in some scenarios. We then demonstrate its utility in several real-world applications: a tomographic reconstruction pipeline, a workflow for the discovery of metal-organic frameworks for carbon capture, and the instrumentation of Dask workflows for provenance tracking and performance analysis.

HPC↗

DOE OSTI2020

Threadsafe Dynamic Neighbor Lists for Monte Carlo Ray Tracing

Monte Carlo (MC) transport codes offer high-fidelity modeling of particle transport physics, but their high computational cost makes them impractical for many applications. For some applications such as multiphysics and depletion that use finely discretized geometries, a large portion of this computational cost is attributable to ray tracing. Neighbor lists are a well-known method for accelerating ray-tracing calculations in a MC code, but despite their prevalence, little work has been published on the details of their implementation. The fine details can have a significant impact on performance, particularly when using shared-memory parallelism. This paper addresses these details of implementation with a discussion of different neighbor list schemes and their impact on software runtime. Performance tests were run by using OpenMC on a pin-cell problem discretized with up to 200 axial regions. The results demonstrate that switching from surface-based to cell-based neighbor lists leads to a 10 faster calculation rate for the most fine discretization. Finally, using a threadsafe shared-memory data structure results in a 20% faster calculation rate versus simple threadprivate neighbor lists. Results here show that a data structure that is contiguous in memory improves performance by only 1% to 2% over noncontiguous linked lists.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

DOE OSTI2025

Thermonuclear Burn in a Multiphysics Code on GPUs

Multiphysics codes links to a library called SINGE for the calculation of thermonuclear (TN) burn rates, but some current multiphysics codes do not attempt to leverage the support for parallel operation that SINGE provides. Our goal is to investigate implementations of the SINGE workflow and analyze how the use of a performance portability layer could reduce run time on CPU archi tectures while also supporting GPU architectures without requiring code modifications. We looked to the Kokkos C++ Performance Portability Ecosystem to implement hardware agnostic parallel patterns.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

DOE OSTI2023

HipBone: A performance-portable graphics processing unit-accelerated C++ version of the NekBone benchmark

We present hipBone, an open-source performance-portable proxy application for the Nek5000 (and NekRS) computational fluid dynamics applications. HipBone is a fully GPU-accelerated C++ implementation of the original NekBone CPU proxy application with several novel algorithmic and implementation improvements which optimize its performance on modern fine-grain parallel GPU accelerators. Our optimizations include a conversion to store the degrees of freedom of the problem in assembled form in order to reduce the amount of data moved during the main iteration and a portable implementation of the main Poisson operator kernel. We demonstrate near-roofline performance of the operator kernel on three different modern GPU accelerators from two different vendors. We present a novel algorithm for splitting the application of the Poisson operator on GPUs which aggressively hides MPI communication required for both halo exchange and assembly. Our implementation of nearest-neighbor MPI communication then leverages several different routing algorithms and GPU-Direct RDMA capabilities, when available, which improves scalability of the benchmark. We demonstrate the performance of hipBone on three different clusters housed at Oak Ridge National Laboratory, namely, the Summit supercomputer and the Frontier early-access clusters, Spock and Crusher. Our tests demonstrate both portability across different clusters and very good scaling efficiency, especially on large problems.

Computer Science↗