SEARCH · Search NASA

Results for “task parallelism”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 595 records · Page 33

All sources NASA NTRS DOE OSTI

DOE OSTI2021

Capacity Investment under Bayesian Information Updates at Reporting Periods: Model and Application

We consider capacity addition decisions by a new product manufacturer faced with uncertain technology alternatives. The manufacturing capacity addition and technology development occurs in parallel, with preliminary results from a technology project's success providing valuable information to the manufacturer in adding capacity. We solve a stochastic dynamic program with Bayesian updates to obtain the manufacturer's expected profit‐maximizing capacity investment decision. Our model and applications are motivated by the Critical Materials Institute (CMI) (funded by the Department of Energy), which manages research projects focused on mitigating critical material constraints, vital to renewable energy technologies such as direct‐drive wind turbines, electric vehicles, and energy‐efficient lighting. We capture three unique aspects of the problem: first, the learning from project progress depends on task‐based stochastic outcomes and a project's percent‐done. Second, the underlying technology's profitability is based on a model of competition. Third, we evaluate the impact of progress across a portfolio of projects based on a manufacturer's capacity addition. We develop a heuristic that produces results that are close to optimal and can thus be used for large problem sizes. The managerial insights from an application of our model to CMI projects include: (i) technology projects that report the percent‐done of a project earlier increase expected manufacturer profit; (ii) careful choice of “safe bets,” that is, technologies with low profitability but a high probability of success, can increase expected manufacturer profit; (iii) a portfolio of projects can increase profits significantly over separate project evaluation; and (iv) dynamic management of project resources can increase overall profit.

Vedantam, Aditya↗

DOE OSTI2024

FLARE: field line analysis and reconstruction for 3D boundary plasma modeling

The FLARE code is a magnetic mesh generator that is integrated within a suite of tools for the analysis of the magnetic geometry in toroidal fusion devices. A magnetic mesh is constructed from field line segments and permits fast reconstruction of field lines in 3D boundary plasma codes such as EMC3-EIRENE. Both intrinsically non-axisymmetric configurations (stellarators) and those with symmetry breaking perturbations of an axisymmetric equilibrium (tokamaks) are supported. The code itself is written in Modern Fortran with MPI support for parallel computing, and it incorporates object-oriented programming for the definition of the magnetic field and the material surface geometry. Extended derived types for a number of different magnetohydrodynamic equilibrium and plasma response models are implemented. The core element of FLARE is a field line tracer with adaptive step-size control, and this is integrated into tools for the construction of Poincaré maps and invariant manifolds of X-points. A collection of high-level procedures that generate output files for visualization is build on top of that. The analysis modules are build with Python frontends that facilitate customization of tasks and/or scripting of parameter scans.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

DOE OSTI2022

Enabling machine learning-ready HPC ensembles with Merlin

With the growing complexity of computational and experimental facilities, many scientific researchers are turning to machine learning (ML) techniques to analyze large scale ensemble data. With complexities such as multi-component workflows, heterogeneous machine architectures, parallel file systems, and batch scheduling, care must be taken to facilitate this analysis in a high performance computing (HPC) environment. Here, we present Merlin, a workflow framework to enable large ML-friendly ensembles of scientific HPC simulations. By augmenting traditional HPC with distributed compute technologies, Merlin aims to lower the barrier for scientific subject matter experts to incorporate ML into their analysis. As a producer–consumer workflow model, Merlin enables multi-machine, cross-batch job, dynamically allocated yet persistent workflows capable of utilizing surge-compute resources. Key features of Merlin are a flexible HPC-centric interface, low per-task overhead, multi-tiered fault recovery, and a hierarchical sampling algorithm that allows for $\mathscr{O}$(N) task execution and $\mathscr{O}$(N ln N) task queuing to ensembles of millions of tasks. In addition to Merlin’s design, we test the algorithm’s performance in an HPC center and demonstrate the ability to enqueue 40 million simulations in 100 s, with a 30 millisecond per-task overhead that is independent of ensemble size. Finally, we describe some example applications that Merlin has enabled on leadership-class HPC resources, such as the ML-augmented optimization of nuclear fusion experiments and the calibration of infectious disease models to study the progression of and possible mitigation strategies for COVID-19.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2023

Next-Generation Marine Energy Software Needs Assessment

Over the past decade the marine energy industry has continued to grow and evolve, with new concepts and technologies constantly being pursued. Additionally, the field of computing is vastly different today than it was five or ten years ago. By utilizing advanced software and hardware architectures, like graphics processing units as well as parallelization and high-performance computing resources, software can produce higher quality outputs and a higher volume of outputs. These software and hardware resources can enable the marine energy community to exploit computational advancements from other research fields, which can include machine learning, differentiable programming, and controls co-design. Better integration of existing software and development of potential new software is necessary to take advantage of trends in modern computing and respond to the current and future needs of the marine energy community. In order to better understand the existing marine energy software landscape and industry needs, DOE's Water Power Technologies Office (WPTO) tasked Sandia National Laboratories and the National Renewable Energy Laboratory to update the needs assessment by identifying existing software gaps and software needs, and assisting WPTO in planning the next wave of marine energy software development. The proposed effort involved cataloguing and analyzing the available data on existing software related to marine energy. The marine energy software landscape has vastly changed in the last ten years. There are now nearly 230 different software packages utilized by the marine energy sector, compared to a decade ago when the Cardinal Engineering survey identified approximately 40 software packages. In 2012, the marine energy software landscape was captured in two tables, whereas the current marine energy software landscape required development of a software database to collect and categorize software.

16 TIDAL AND WAVE POWER↗

DOE OSTI2021

Progress on Optimizing Wind Farms and Rotor Designs Using Adjoints

Modern wind plants are increasingly tasked with multiple performance objectives. In addition to designing plants that maximize power output and minimize the levelized cost of energy (LCOE), the design and operation of wind plants is increasingly influenced by challenges regarding grid integration of variable generation renewables. This places a growing emphasis on making wind plants more controllable and predictable. WindSE is a Reynolds-averaged Navier-Stokes (RANS) model designed around analytical gradient and adjoint methods, with the ability to capture terrain-induced effects, as shown in Figure 1. The recent addition of an unsteady solver with an actuator line method (ALM) and ongoing work to enable massively parallel optimizations gives it a unique niche to explore coupled plant-level controls and design problems. This code is an open source python package built on the FEniCS framework that utilizes fast, parallel PETSc solvers to model fluid flow throughout wind-farm scale domains. Two recent studies performed using WindSE demonstrate the capability to optimize under a wide variety of flow conditions and objective functions. In the first, we present an optimization focused on modifying the layout of a wind farm with a fixed number of turbines for maximum total power output [1]. This study highlights the ability to quickly perform simulations using the steady Navier-Stokes solver combined with rotors represented as actuator disks while also stressing the importance of capturing terrain-induced effects. Gradient-based optimization using the RANS equations is viable due to the inclusion of efficiently computed adjoint derivatives. We interpret the physical results of the optimal layout and also discuss the computational cost of scaling to larger problems. In the second study, we present the capabilities of the unsteady Navier-Stokes solver, where rotor-blade profiles represented by actuator lines are optimized to enhance wake steering effects and overall power production [2]. We quantify the wind plant performance gains obtained from this type of simultaneous control co-design optimization as compared to optimizing the blade design and yaw independently. Figure 2 shows the differences between a baseline two-turbine system and an optimized system where we fine-tune the blade chord profile. Results and challenges from each study are quickly summarized and used to motivate the current development efforts within WindSE. Current and future work is focused on enabling higher-resolution studies with more degrees of freedom through parallelization of both the simulation and optimization algorithms. We present benchmarking results to show that WindSE performs well in both weak- and strong-scaling tests and further demonstrate that the optimizer obtains the same convergence rates in both shared- and distributed-memory environments. Using larger wind farms, we can study deep-array effects within an optimization context, allowing the use of objective functions that have been previously unstudied. As an example, we present ongoing work on a blockage metric which characterizes the loss of available kinetic energy due to wake effects from multiple upstream turbines.

adjoint optimization↗

DOE OSTI2016

On the performance of exponential integrators for problems in magnetohydrodynamics

Exponential integrators have been introduced as an efficient alternative to explicit and implicit methods for integrating large stiff systems of differential equations. Over the past decades these methods have been studied theoretically and their performance was evaluated using a range of test problems. While the results of these investigations showed that exponential integrators can provide significant computational savings, the research on validating this hypothesis for large scale systems and understanding what classes of problems can particularly benefit from the use of the new techniques is in its initial stages. Resistive magnetohydrodynamic (MHD) modeling is widely used in studying large scale behavior of laboratory and astrophysical plasmas. In many problems numerical solution of MHD equations is a challenging task due to the temporal stiffness of this system in the parameter regimes of interest. In this paper we evaluate the performance of exponential integrators on large MHD problems and compare them to a state-of-the-art implicit time integrator. Both the variable and constant time step exponential methods of EPIRK-type are used to simulate magnetic reconnection and the Kevin–Helmholtz instability in plasma. Performance of these methods, which are part of the EPIC software package, is compared to the variable time step variable order BDF scheme included in the CVODE (part of SUNDIALS) library. We study performance of the methods on parallel architectures and with respect to magnitudes of important parameters such as Reynolds, Lundquist, and Prandtl numbers. We find that the exponential integrators provide superior or equal performance in most circumstances and conclude that further development of exponential methods for MHD problems is warranted and can lead to significant computational advantages for large scale stiff systems of differential equations such as MHD.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2026

matsim-agents v1.0

matsim-agents is a multi-agent AI framework for atomistic materials simulation and discovery. It orchestrates large language models (LLMs), machine-learned interatomic potentials (MLIPs), and DFT codes into a single agentic loop running on laptops and DOE leadership-class supercomputers. MULTI-AGENT ORCHESTRATION A LangGraph state machine with three nodes: a Planner that converts a natural-language research objective into structured tasks; an Executor that dispatches atomistic tools and loops until the queue is empty; and an Analyst that summarizes results into a human-readable report. State is checkpointed after every step and human-in-the-loop gates can be inserted at any edge. HYPOTHESIS-DRIVEN DISCOVERY CHAT An interactive REPL (matsim-agents chat) that couples LLM dialogue with atomistic simulation. Chemical formulas are automatically detected in conversation turns and trigger a full crystal-phase exploration: structure generation → relaxation → stability scoring → result injection back into the conversation, creating a closed hypothesis-refinement loop. CRYSTAL PHASE ENUMERATION Given a composition, the phase explorer enumerates prototypes by stoichiometry: elemental (fcc/bcc/hcp/sc/diamond), binary 1:1 (rocksalt/CsCl/zincblende/ wurtzite/fluorite/rutile), ternary 1:1:3 (cubic perovskite), ternary 1:2:4 (perovskite + spinel), quaternary 1:1:2:6 (Fm-3m double perovskite). 2-D prototypes (graphene, h-BN, MoS2 2H/1T) and multilayer stacking are also supported via --include-2d and --num-layers. SUPERCELL GENERATION AND SITE DECORATION Auto-tiling to a minimum atom count (--min-atoms), explicit NxNxN tiling (--supercell), symmetry-distinct site decorations (--n-orderings), and isotropic lattice-scale sweeps (--lattice-scales) for volume bracketing. MLFF RELAXATION AND STABILITY SCORING HydraGNN (multi-headed GNN) drives structure relaxation via ASE with FIRE, BFGS, or BFGSLineSearch. Stability output: delta-E/atom ranking across phases and a max-residual-force dynamical-stability proxy. Other MLIPs (MACE, NequIP, Orb) can be plugged in through the same interface. DFT BACKENDS Quantum ESPRESSO pw.x and VASP 6.6 are first-class labellers. Both have validated GPU builds and SLURM/PBS launchers for three DOE platforms: Frontier (AMD MI250X, ROCm), Aurora (Intel PVC, oneAPI), Perlmutter (NVIDIA A100, CUDA). QE produces ~100 binaries (pw.x, ph.x, epw.x, ...). VASP supports scf, relax, vc-relax, and vc-relax-shape run types. ACTIVE-LEARNING LOOP matsim-agents al run CONFIG.yaml drives an iterative HydraGNN-DFT loop: MD generates candidates → ensemble/MC-dropout uncertainty selects the most informative → DFT labels them in parallel inside one allocation → dataset grows → HydraGNN retrains → repeat. DFT backend is a single YAML toggle (dft.backend: vasp | qe). LLM-generated seed structures are supported (no curated POSCAR library needed). Config uses ${VAR}, ${VAR:-default}, ${VAR:?msg} shell-style substitution for cross-user/cross-site portability. LLM BACKENDS Ollama (local, default), vLLM (HPC multi-GPU serving), OpenAI, Anthropic, HuggingFace Transformers+Accelerate. Selected at runtime via flag or env var with no code changes. HPC PORTABILITY Same Python entry points run on Frontier (ROCm 7.2), Aurora (oneAPI), and Perlmutter (CUDA 12). DFT and ML stacks are never co-loaded in the same shell; they couple through the scheduler and filesystem. Advanced multi-node launchers (serve, discovery-chat, single-relaxation, active-learning, QE warm-start) are provided for all three platforms. CODABENCH COMPETITION BUNDLE A self-contained benchmark: 159 atomistic test structures across 11 material classes, 5 tasks (formation energy, forces, ML relaxation, AI-DFT relaxation, phase stability ranking), public/private leaderboard split (30/70), and four ready-to-run baselines: MACE-MP-0, HydraGNN, UMA, AllScAIP.

Lupo Pasini, Massimiliano [Oak Ridge National Labo↗

DOE OSTI2015

Exploiting communication concurrency on high performance computing systems

Although logically available, applications may not exploit enough instantaneous communication concurrency to maximize hardware utilization on HPC systems. This is exacerbated in hybrid programming models such as SPMD+OpenMP. We present the design of a "multi-threaded" runtime able to transparently increase the instantaneous network concurrency and to provide near saturation bandwidth, independent of the application configuration and dynamic behavior. The runtime forwards communication requests from application level tasks to multiple communication servers. Our techniques alleviate the need for spatial and temporal application level message concurrency optimizations. Experimental results show improved message throughput and bandwidth by as much as 150% for 4KB bytes messages on InfiniBand and by as much as 120% for 4KB byte messages on Cray Aries. For more complex operations such as all-to-all collectives, we observe as much as 30% speedup. This translates into 23% speedup on 12,288 cores for a NAS FT implemented using FFTW. We also observe as much as 76% speedup on 1,500 cores for an already optimized UPC+OpenMP geometric multigrid application using hybrid parallelism.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2016

An Adaptable Seismic Data Format

We present ASDF, the Adaptable Seismic Data Format, a modern and practical data format for all branches of seismology and beyond. The growing volume of freely available data coupled with ever expanding computational power opens avenues to tackle larger and more complex problems. Current bottlenecks include inefficient resource usage and insufficient data organization. Properly scaling a problem requires the resolution of both these challenges, and existing data formats are no longer up to the task. ASDF stores any number of synthetic, processed or unaltered waveforms in a single file. A key improvement compared to existing formats is the inclusion of comprehensive meta information, such as event or station information, in the same file. Additionally, it is also usable for any non-waveform data, for example, cross-correlations, adjoint sources or receiver functions. Last but not least, full provenance information can be stored alongside each item of data, thereby enhancing reproducibility and accountability. Any data set in our proposed format is self-describing and can be readily exchanged with others, facilitating collaboration. The utilization of the HDF5 container format grants efficient and parallel I/O operations, integrated compression algorithms and check sums to guard against data corruption. To not reinvent the wheel and to build upon past developments, we use existing standards like QuakeML, StationXML, W3C PROV and HDF5 wherever feasible. Usability and tool support are crucial for any new format to gain acceptance. We developed mature C/Fortran and Python based APIs coupling ASDF to the widely used SPECFEM3D_GLOBE and ObsPy toolkits.

58 GEOSCIENCES↗

DOE OSTI2024

Prompt Phrase Ordering Using Large Language Models in HPC: Evaluating Prompt Sensitivity

Large language models (LLMs) have demonstrated effective performance in domain-specific tasks, often requiring a well-designed prompt to guide their responses. However, optimizing the right prompt is challenging due to prompt sensitivity—the phenomenon where small changes in the prompt can lead to significant variations in performance. In this study, we evaluate prompt performance by examining all permutations of independent phrases to investigate prompt sensitivity and robustness. We used two datasets: the GSM8k dataset, which assesses mathematical reasoning, and a custom template prompt for summarizing database metadata. Our goal was to evaluate the performance across all permutations of a sequence of prompt phrases. The study was conducted using the llama3-instruct- 7B model hosted on Ollama, with computations parallelized in a high-performance computing environment. By comparing the average index of phrases in the best and worst-performing prompts, we found that the order of independent phrases within a prompt significantly impacts LLM performance. Additionally, we used Hamming distance to assess changes between phrase orderings, concluding that prompt modifications can dramatically affect scores, often by almost random chance. These findings support existing research on prompt sensitivity. We discuss the challenges of prompt optimization, noting that altering phrases in a successful prompt does not always result in another successful prompt.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2015

Developing Uintah's Runtime System for Forthcoming Architectures

The need to solve ever-more complex science and engineering problems on computer architectures that are rapidly evolving poses considerable challenges for modern simulation frameworks and packages. While the decomposition of software into a programming model that generates a task-graph for execution by a runtime system make portability and performance possible, it also imposes considerable demands on that runtime system. Here, the challenge of meeting those demands for near-term and longer term systems in the context of the Uintah framework is considered in key areas such as support for data structures, tasks on heterogeneous architectures, performance portability, power management and designing for resilience by the use of replication based upon adaptive mesh refinement processes is addressed. Examples of techniques in these areas are used to illustrate performance improvements obtained for present large-scale systems and the needs of proposed systems considered.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2017

Benefits of Integration of Aerojet Rocketdyne and RTI Advanced Gasification Technologies for Hydrogen-Rich Syngas Production

The key objective for this project was to evaluate the potential to achieve substantial reductions in the production cost of H 2 -rich syngas via coal gasification with near-zero emissions due to the cumulative and synergistic benefits realized when multiple advanced technologies are integrated into the overall conversion process. In this project, Aerojet Rocketdyne’s (AR’s) advanced gasification technology (currently being offered as R-GAS™) and RTI International’s (RTI’s) advanced warm syngas cleanup technologies were evaluated via a number of comparative techno-economic case studies. AR’s advanced gasification technology consists of a dry solids pump and a compact gasifier system. Based on the unique design of this gasifier, it has been shown to reduce the capital cost of the gasification block by between 40 and 50%. At the start of this project, actual experimental work had been demonstrated through pilot plant systems for both the gasifier and dry solids pump. RTI’s advanced warm syngas cleanup technologies consist primarily of RTI’s Warm Gas Desulfurization Process (WDP) technology, which effectively allows decoupling of the sulfur and CO 2 removal allowing for more flexibility in the selection of the CO 2 removal technology, plus associated advanced technologies for direct sulfur recovery and water gas shift (WGS). WDP has been demonstrated at pre-commercial scale using an activated amine carbon dioxide recovery process which would not have been possible if a majority of the sulfur had not been removed from the syngas by WDP. This pre-commercial demonstration of RTI’s advanced warm syngas cleanup system was conducted in parallel to the activities on this project. The technical data and cost information from this pre-commercial demonstration were extensively used in this project during the techno-economic analysis. With this project, both of RTI’s advanced WGS technologies were investigated. Because RT’s advanced fixed-bed WGS (AFWGS) process was successfully implemented in the WDP pre-commercial demonstration test mentioned above, this technology was used as part of RTI’s advanced warm syngas technology package for the techno-economic analyses for this project. RTI’s advanced transport-reactor-based WGS (ATWGS) process was still conceptual at the start of this project, but one of the tasks for this project was to evaluate the technical feasibility of this technology. In each of the three application-based comparison studies conducted as part of this project, the reference case was based on an existing Department of Energy National Energy Technology Laboratory (DOE/NETL) system study. Each of these references cases used existing commercial technology and the system resulted in > 90% carbon capture. In the comparison studies for the use of the hydrogen-rich syngas generated in either an Integrated Gasification Combined Cycle (IGCC) or a Coal-to-Methanol (CTM) plant, the comparison cases consisted of the reference case, a case with the integration of each individual advanced technology (either AR or RTI), and finally a case with the integration of all the advanced technologies (AR and RTI combined). In the Coal-to-Liquids (CTL) comparison study, the comparison study consisted of only three cases, which included a reference case, a case with just RTI’s advanced syngas cleaning technology, and a case with AR’s and RTI’s advanced technologies. The results from these comparison studies showed that the integration of the advanced technologies did result in substantial benefits, and by far the greatest benefits were achieved for cases integrating all the advanced technologies. For the IGCC study, the fully integrated case resulted in a 1.4% net efficiency improvement, an 18% reduction in capital cost per kW of capacity, a 12% reduction in the operating cost per kWh, and a 75–79% reduction in sulfur emissions. For the CTM case, the fully integrated plant resulted in a 22% reduction in capital cost, a 13% reduction in operating costs, a > 99% net reduction in sulfur emissions, and a reduction of 13–15% in CO 2 emissions. Because the capital cost represents over 60% of the methanol Required Selling Price (RSP), the significant reduction in the capital cost for the advanced technology case resulted in an 18% reduction in methanol RSP. For the CTL case, the fully integrated plant resulted in a 16% reduction in capital cost, which represented a 13% reduction in diesel RSP. Finally, the technical feasibility analysis of RTI’s ATWGS process demonstrated that a fluid-bed catalyst with sufficient attrition resistance and WGS activity could be made and that the process achieved about a 24% reduction in capital cost compared to a conventional fixed-bed commercial process.

08 HYDROGEN↗

DOE OSTI2017

NSR&D FY17 Report: CartaBlanca Capability Enhancements

Over the last several years, particle technology in the CartaBlanca code has been matured and has been successfully applied to a wide variety of physical problems. It has been shown that the particle methods, especially Los Alamos's dual domain material point method, is capable of computing many problems involves complex physics, chemistries accompanied by large material deformations, where the traditional finite element or Eulerian method encounter significant difficulties. In FY17, the CartaBlanca code has been enhanced with physical models and numerical algorithms. We started out to compute penetration and HE safety problems. Most of the year we focused on the TEPLA model improvement testing against the sweeping wave experiment by Gray et al., because it was found that pore growth and material failure are essentially important for our tasks and needed to be understood for modeling the penetration and the can experiments efficiently. We extended the TEPLA mode from the point view of ensemble phase average to include the effects of nite deformation. It is shown that the assumed pore growth model in TEPLA is actually an exact result from the theory. Alone this line, we then generalized the model to include finite deformations to consider nonlinear dynamics of large deformation. The interaction between the HE product gas and the solid metal is based on the multi-velocity formation. Our preliminary numerical results suggest good agreement between the experiment and the numerical results, pending further verification. To improve the parallel processing capabilities of the CartaBlanca code, we are actively working with the Next Generation Code (NGC) project to rewrite selected packages using C++. This work is expected to continue in the following years. This effort also makes the particle technology developed with CartaBlanca project available to other part of the laboratory. Working with the NGC project and rewriting some parts of the code also given us an opportunity to improve our numerical implementations of the method and to take advantage of recently advances in the numerical methods, such as multiscale algorithms.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2026

Surrogate Neural Architecture Codesign Package (SNAC-Pack)

Neural architecture search (NAS) is a powerful approach for automating model design, but existing methods often optimize for accuracy alone or rely on proxy metrics such as bit operations (BOPs) that correlate poorly with hardware cost. This gap is particularly large for FPGA deployment, where cost is dominated by a multi-dimensional budget of lookup tables, DSPs, flip-flops, BRAM, and latency. We present the Surrogate Neural Architecture Codesign Package (SNAC-Pack), an open-source AutoML framework for hardware-aware neural architecture codesign and end-to-end FPGA deployment. SNAC-Pack runs a multi-objective global search with Optuna and NSGA-II, loading trials to a shared SQLite store that enables parallel workers across compute nodes. A hardware surrogate model outputs per-trial resource and latency estimates, avoiding the synthesis cost that would otherwise dominate the search loop. A local search stage then applies quantization-aware training (QAT) together with iterative magnitude pruning in a combined compression loop, after which the final model is synthesized to FPGA firmware via the hls4ml Python library. A YAML configuration and an optional agentic frontend let users run the pipeline on new datasets without modifying the framework. We demonstrate SNAC-Pack on jet classification at the Large Hadron Collider and superconducting qubit readout, discovering compact architectures that match or exceed strong baselines on the task metric while reducing FPGA resource utilization and, in the qubit readout case, reducing the design space exploration process from months of manual fine-tuning to hours of automated search.

Weitz, Jason [UC, San Diego]↗

DOE OSTI2017

Traleika Glacier: A hardware-software co-designed approach to exascale computing

The move from current petascale machines to future exascale machines will need both hardware improvements and software changes. Hardware will need to evolve to focus primarily on features that lower energy consumption: near-threshold voltage operation, fine-grained power and clock management and heterogeneity. Software will also need to evolve and be able to express more parallelism, become more dynamic and adaptable in order to be able to operate on a much more variable hardware. Here, we present Traleika Glacier, an effort that seeks to evaluate radical design changes to meet the constraints, both in terms of power and cost, of exascale computing. The salient features of the hardware design presented in the work include a) a use of heterogeneous cores, b) a redesign of the memory system that centers around hierarchical scratchpads and a global address space, c) the hardware acceleration of certain memory and network operations through specialized engines and, d) very fine-grained control and monitoring capabilities. On the software side, we describe a task-based runtime system, the Open Community Runtime (OCR) which aims to express a wide range of higher-level programming models with a very limited set of core concepts: event-driven tasks for computation, events for synchronization and relocatable data-blocks for data management.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2016

A Fault-oblivious Extreme-scale Execution Environment

Exascale computing systems will provide a thousand-fold increase in parallelism and a proportional increase in failure rate relative to today's machines. Systems software for exascale machines must provide the infrastructure to support existing applications while simultaneously enabling efficient execution of new programming models that naturally express dynamic, adaptive, irregular computation; coupled simulations; and massive data analysis in a highly unreliable hardware environment with billions of threads of execution. We propose a new approach to the data and work distribution model provided by system software based on the unifying formalism of an abstract file system. The proposed hierarchical data model provides simple, familiar visibility and access to data structures through the file system hierarchy, while providing fault tolerance through selective redundancy. The hierarchical task model features work queues whose form and organization are represented as file system objects. Data and work are both first class entities. By exposing the relationships between data and work to the runtime system, information is available to optimize execution time and provide fault tolerance. The data distribution scheme provides replication (where desirable and possible) for fault tolerance and efficiency, and it is hierarchical to make it possible to take advantage of locality. The user, tools, and applications, including legacy applications, can interface with the data, work queues, and one another through the abstract file model. This runtime environment will provide multiple interfaces to support traditional Message Passing Interface applications, languages developed under DARPA's High Productivity Computing Systems program, as well as other, experimental programming models. We will validate our runtime system with pilot codes on existing platforms and will use simulation to validate for exascale-class platforms. In this final report, we summarize research results from the work done at the Ohio State University towards the larger goals of the project listed above.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2020

Enabling Highly Efficient Capsule Networks Processing Through A PIM-Based Architecture Design

In recent years, the CNNs have achieved great successes in the image processing tasks, e.g., image recognition and object detection. Unfortunately, traditional CNN's classication is found to be easily misled by increasingly complex image features due to the usage of pooling operations, hence unable to preserve accurate position and pose information of the objects. To address this challenge, a novel neural network structure called Capsule Network has been proposed, which introduces equivariance through capsules to signicantly enhance the learning ability for image segmentation and object detection. Due to its requirement of performing a high volume of matrix operations, CapsNets have been generally accelerated on modern GPU platforms that provide highly optimized software library for common deep learning tasks. However, based on our performance characterization on modern GPUs, CapsNets exhibit low effciency due to the special program and execution features of their routing procedure, including massive unshareable intermediate variables and intensive syn- chronizations, which are very dicult to optimize at software level. To address these challenges, we propose a hybrid computing architecture design named PIM-CapsNet. It preserves GPU's on-chip computing capability for accelerating CNN types of layers in CapsNet, while pipelining with an off-chip in-memory acceleration solution that effectively tackles routing procedure's ineffciency by leveraging the processing-in-memory capability of today's 3D stacked memory. Using routing procedure's inherent parallellization feature, our design enables hierarchical improvements on CapsNet inference effciency through minimizing data movement and maximizing parallel processing in memory. Evaluation results demonstrate that our proposed design can achieve substantial improvement on both performance and energy savings for CapsNet inference, with almost zero accuracy loss. The results also suggest good performance scalability in optimizing the routing procedure with increasing network size.

Zhang, Xingyao↗

DOE OSTI2015

Stable large-scale solver for Ginzburg–Landau equations for superconductors

Understanding the interaction of vortices with inclusions in type-II superconductors is a major outstanding challenge both for fundamental science and energy applications. At application-relevant scales, the long-range interactions between a dense configuration of vortices and the dependence of their behavior on external parameters, such as temperature and an applied magnetic field, are all important to the net response of the superconductor. Capturing these features, in general, precludes analytical description of vortex dynamics and has also made numerical simulation prohibitively expensive. Here we report on a highly optimized iterative implicit solver for the time-dependent Ginzburg-Landau equations suitable for investigations of type-II superconductors on massively parallel architectures. Its main purpose is to study vortex dynamics in disordered or geometrically confined mesoscopic systems. In this work, we present the discretization and time integration scheme in detail for two types of boundary conditions. We describe the necessary conditions for a stable and physically accurate integration of the equations of motion. Using an inclusion pattern generator, we can simulate complex pinning landscapes and the effect of geometric confinement. We show that our algorithm, implemented on a GPU, can provide static and dynamic solutions of the Ginzburg-Landau equations for mesoscopically large systems over thousands of time steps in a matter of hours. Using our formulation, studying scientifically-relevant problems is a computationally reasonable task.

Sadovskyy, I. A.↗