SEARCH · Engineering Papers

Results for “Parallel processing”

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 577 records · Page 32

All sources NASA NTRS DOE OSTI

DOE OSTI2024

Compiling Quantum Circuits for Dynamically Field-Programmable Neutral Atoms Array Processors

Dynamically field-programmable qubit arrays (DPQA) have recently emerged as a promising platform for quantum information processing. In DPQA, atomic qubits are selectively loaded into arrays of optical traps that can be reconfigured during the computation itself. Leveraging qubit transport and parallel, entangling quantum operations, different pairs of qubits, even those initially far away, can be entangled at different stages of the quantum program execution. Such reconfigurability and non-local connectivity present new challenges for compilation, especially in the layout synthesis step which places and routes the qubits and schedules the gates. In this paper, we consider a DPQA architecture that contains multiple arrays and supports 2D array movements, representing cutting-edge experimental platforms. Within this architecture, we discretize the state space and formulate layout synthesis as a satisfiability modulo theories problem, which can be solved by existing solvers optimally in terms of circuit depth. For a set of benchmark circuits generated by random graphs with complex connectivities, our compiler OLSQ-DPQA reduces the number of two-qubit entangling gates on small problem instances by 1.7x compared to optimal compilation results on a fixed planar architecture. To further improve scalability and practicality of the method, we introduce a greedy heuristic inspired by the iterative peeling approach in classical integrated circuit routing. Using a hybrid approach that combined the greedy and optimal methods, we demonstrate that our DPQA-based compiled circuits feature reduced scaling overhead compared to a grid fixed architecture, resulting in 5.1X less two-qubit gates for 90 qubit quantum circuits. These methods enable programmable, complex quantum circuits with neutral atom quantum computers, as well as informing both future compilers and future hardware choices.

Physics↗

DOE OSTI2022

Block-structured, equal-workload, multi-grid-nesting interface for the Boussinesq wave model FUNWAVE-TVD (Total Variation Diminishing)

Abstract. We describe the development of a block-structured, equal-CPU-load (central processing unit), multi-grid-nesting interface for the Boussinesq wave model FUNWAVE-TVD (Fully Nonlinear Boussinesq Wave Model with Total Variation Diminishing Solver). The new model framework does not interfere with the core solver, and thus the core program, FUNWAVE-TVD, is still a standalone model used for a single grid. The nesting interface manages the time sequencing and two-way nesting processes between the parent grid and child grid with grid refinement in a hierarchical manner. Workload balance in the MPI-based (message passing interface) parallelization is handled by an equal-load scheme. A strategy of shared array allocation is applied for data management that allows for a large number of nested grids without creating additional memory allocations. Four model tests are conducted to verify the nesting algorithm with assessments of model accuracy and the robustness in the application in modeling transoceanic tsunamis and coastal effects.

Choi, Young-Kwang↗

DOE OSTI2023

SERGHEI (SERGHEI-SWE) v1.0: a performance-portable high-performance parallel-computing shallow-water solver for hydrology and environmental hydraulics

The Simulation EnviRonment for Geomorphology, Hydrodynamics, and Ecohydrology in Integrated form (SERGHEI) is a multi-dimensional, multi-domain, and multi-physics model framework for environmental and landscape simulation, designed with an outlook towards Earth system modelling. At the core of SERGHEI's innovation is its performance-portable high-performance parallel-computing (HPC) implementation, built from scratch on the Kokkos portability layer, allowing SERGHEI to be deployed, in a performance-portable fashion, in graphics processing unit (GPU)-based heterogeneous systems. In this work, we explore combinations of MPI and Kokkos using OpenMP and CUDA backends. In this contribution, we introduce the SERGHEI model framework and present with detail its first operational module for solving shallow-water equations (SERGHEI-SWE) and its HPC implementation. This module is designed to be applicable to hydrological and environmental problems including flooding and runoff generation, with an outlook towards Earth system modelling. Its applicability is demonstrated by testing several well-known benchmarks and large-scale problems, for which SERGHEI-SWE achieves excellent results for the different types of shallow-water problems. Finally, SERGHEI-SWE scalability and performance portability is demonstrated and evaluated on several TOP500 HPC systems, with very good scaling in the range of over 20 000 CPUs and up to 256 state-of-the art GPUs.

58 GEOSCIENCES↗

DOE OSTI2021

ParFlow

ParFlow is an open-source, modular, parallel watershed flow model. It includes fully-integrated overland flow, the ability to simulate complex topography, geology and heterogeneity and coupled land-surface processes including the land-energy budget, biogeochemistry and snow (via CLM). It is multi-platform and runs with a common I/O structure from laptop to supercomputer. ParFlow is the result of a long, multi-institutional development history and is now a collaborative effort between CSM, LLNL, UniBonn and UCB. ParFlow has been coupled to the mesoscale, meteorological code ARPS and the NCAR code WRF.

Smith, Steven↗

DOE OSTI2021

ParFlow

Smith, Steven↗

DOE OSTI2024

Boosting RDataFrame performance with transparent bulk event processing

RDataFrame is ROOT’s high-level interface for Python and C++ data analysis. Since it first became available, RDataFrame adoption has grown steadily and it is now poised to be a major component of analysis software pipelines for LHC Run 3 and beyond. Thanks to its design inspired by declarative programming principles, RDataFrame enables the development of highperformance, highly parallel analyses without requiring expert knowledge of multi-threading and I/O: user logic is expressed in terms of self-contained, small computation kernels tied together by a high-level API. This design completely decouples analysis logic from its actual execution, and opens several interesting avenues for workflow optimization. In particular, in this work we explore the benefits of moving internal data processing from an event-by-event to a bulkby-bulk loop. This refactoring dramatically reduces the framework’s runtime overheads; in collaboration with the I/O layer it improves data access patterns; it exposes information that optimizing compilers might use to auto-vectorize the invocation of user-defined computations; finally, while existing user-facing interfaces remain unaffected, it becomes possible to additionally offer interfaces that explicitly expose bulks of events, useful e.g. for the injection of GPU kernels into the analysis workflow. In order to inform similar future R&D, design challenges will be presented, as well as an investigation of the relevant timememory trade-off backed by novel performance benchmarks.

Guiraud, Enrico↗

DOE OSTI2023

Reinforcement Learning for Load-balanced Parallel Particle Tracing

We explore an online reinforcement learning (RL) paradigm to dynamically optimize parallel particle tracing performance in distributed-memory systems. Our method combines three novel components: (1) a work donation algorithm, (2) a high-order workload estimation model, and (3) a communication cost model. First, we design an RL-based work donation algorithm. Our algorithm monitors workloads of processes and creates RL agents to donate data blocks and particles from high-workload processes to low-workload processes to minimize program execution time. The agents learn the donation strategy on the fly based on reward and cost functions designed to consider processes' workload changes and data transfer costs of donation actions. Second, we propose a workload estimation model, helping RL agents estimate the workload distribution of processes in future computations. Third, we design a communication cost model that considers both block and particle data exchange costs, helping RL agents make effective decisions with minimized communication costs. We demonstrate that our algorithm adapts to different flow behaviors in large-scale fluid dynamics, ocean, and weather simulation data. Our algorithm improves parallel particle tracing performance in terms of parallel efficiency, load balance, and costs of I/O and communication for evaluations with up to 16,384 processors.

Distributed and parallel particle tracing↗

DOE OSTI2021

Development of Algal Biomass Yield Improvements in an Integrated Process (Final Report)

This project built on the success of Global Algae Innovations Algae Biomass Yield Phase 1 project to accelerate the commercialization of algal biofuels through development of an integrated, economical, photosynthetic, open raceway system to produce algal oil. Two parallel pathways to a biofuel were investigated. In the algal crude oil pathway, highlighted in blue, the dewatered algal biomass slurry is used as a feed to hydrothermal liquefaction to produce an algae crude oil and recycle aqueous stream. In the algal lipid oil pathway, highlighted in green, the algal biomass slurry is dried, and the oil is extracted to produce an algal lipid oil biofuel intermediate and a high protein algae meal co-product. Upgrading to drop-in fuels has been demonstrated for the biofuel intermediates in both pathways. Since the algal lipid oil pathway requires lipid accumulation, the productivity is generally lower than the algal crude oil pathway, but the required productivity for economical algal biofuel production is also lower because the coproduct value is greater. The outdoor cultivation was performed at the Kauai Algae Farm. The project team has expertise across the full breadth strain development, advanced algal cultivation, open raceway contamination control, CO 2 supply, harvesting, dewatering, extraction, hydrothermal liquefaction, technoeconomic analysis and life cycle analysis: Hildebrand laboratory at Scripps Institution of Oceanography – diatom strain development. Mayfield laboratory at UCSD - green algae strain development. TSD Management Associates- CO 2 supply, algal harvesting and dewatering, and algal drying and extraction. Qualitas - large-scale test site for algal harvesting technology. Texas A&M University - algal drying and extraction technology. Pacific Northwest National Laboratory - hydrothermal liquefaction technology. GE Water and Power - membrane technology. National Renewable Energy Laboratory - algal techno-economic analysis. Phase 1 of the project resulted in tremendous productivity and pre-processing improvements in an integrated, large-scale, low-cost cultivation and pre-processing process that moved algal technology closer to economic viability for biofuels than ever before. Phase 2 accomplishments further facilitate development of a commercial algal biofuel industry by adding: several top performing strains and strain development tools including breeding; open-pond cultivation innovations that achieved a 30% improvement in overall productivity, an 83% reduction in cost, and a 95% reduction in energy use relative to conventional technology; further improvements in harvesting that led to the Zobi harvester ® as a universal, economical, low energy, commercially available microalgae harvesting solution; breakthroughs in drying and extraction including a new low energy drying process and several new low energy extraction unit operations that together with the Zobi harvester ® achieve a 90% reduction in cost and 97% reduction in energy use relative to conventional technology. These improvements are sufficient to enable economical production of algae biofuel and protein meal co-product if the process were scaled up to a commercial-scale 5,000-acre algae farm.

09 BIOMASS FUELS↗

DOE OSTI2024

Oblique instability of quasi-parallel whistler waves in the presence of cold and warm electron populations

Whistler waves propagating nearly parallel to the ambient magnetic field experience a nonlinear instability due to transverse currents when the background plasma has a population of sufficiently low energy electrons. Intriguingly, this nonlinear process may generate oblique electrostatic waves, including whistlers near the resonance cone with properties resembling oblique chorus waves in the Earth’s magnetosphere. Focusing on the generation of oblique whistlers, earlier analysis of the instability is extended here to the case where low-energy background plasma consists of both a “cold” population with energy of a few eV and a “warm” electron component with energy of the order of 100 eV. This is motivated by spacecraft observations in the Earth’s magnetosphere where oblique chorus waves were shown to interact resonantly with the warm electrons. The main new results are: 1) the instability producing oblique electrostatic waves is sensitive to the shape of the electron distribution at low energies. In the whistler range of frequencies, two distinct peaks in the growth rate are typically present for the model considered: a peak associated with the warm electron population at relatively low wavenumbers and a peak associated with the cold electron population at relatively high wavenumbers; 2) overall, the instability producing oblique whistler waves near the resonance cone persists (with a reduced growth rate) even in the cases where the temperature of the cold population is relatively high, including cases where cold population is absent and only the warm population is included; 3) particle-in-cell simulations show that the instability leads to heating of the background plasma and formation of characteristic plateau and beam features in the parallel electron distribution function in the range of energies resonant with the instability. The plateau/beam features have been previously detected in spacecraft observations of oblique chorus waves. However, they have been attributed to external sources and have been proposed to be the mechanism generating oblique chorus. In the present scenario, the causality link is reversed and the instability generating oblique whistler waves is shown to be a possible mechanism for formation of the plateau and beam features.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2022

Coincident Capture through Post-processing PTRAC [Slides]

This presentation discusses the new PTRAC capabilities and workflows. The PTRAC capability in MCNP6.3 has seen a massive overhaul since MCNP6.2. The new HDF5 file format allows for both MPI- and thread-based parallelism. MCNPTools has been updated to handle the new HDF5 PTRAC format and is now open sourced on GitHub. Built-in capabilities, such as the pulse-height tally coincident capture special treatment, can largely be replicated through separate postprocessing scripts that leverage both PTRAC and MCNPTools. This allows for greater flexibility in user-specified and controlled detector response functionality, ultimately using the MCNP code for what it is best at (i.e., particle transport).

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

DOE OSTI2021

Revisiting Huffman Coding: Toward Extreme Performance on Modern GPU Architectures

Today's high-performance computing (HPC) applications are producing vast volumes of data, which are challenging to store and transfer efficiently during the execution, such that data compression is becoming a critical technique to mitigate the storage burden and data movement cost. Huffman coding is arguably the most efficient Entropy coding algorithm in information theory, such that it could be found as a fundamental step in many modern compression algorithms such as DEFLATE. On the other hand, today's HPC applications are more and more relying on the accelerators such as GPU on supercomputers, while Huffman encoding suffers from low throughput on GPUs, resulting in a significant bottleneck in the entire data processing. In this paper, we propose and implement an efficient Huffman encoding approach based on modern GPU architectures, which addresses two key challenges: (1) how to parallelize the entire Huffman encoding algorithm, including codebook construction, and (2) how to fully utilize the high memory-bandwidth feature of modern GPU architectures. The detailed contribution is fourfold. (1) We develop an efficient parallel codebook construction on GPUs that scales effectively with the number of input symbols. (2) We propose a novel reduction based encoding scheme that can efficiently merge the codewords on GPUs. (3) We optimize the overall GPU performance by leveraging the state-of-the-art CUDA APIs such as Cooperative Groups. (4) We evaluate our Huffman encoder thoroughly using six real-world application datasets on two advanced GPUs and compare with our implemented multi-threaded Huffman encoder. Experiments show that our solution can improve the encoding throughput by up to 5.0x and 6.8x on NVIDIA RTX 5000 and V100, respectively, over the state-of-the-art GPU Huffman encoder, and by up to 3.3x over the multi-thread encoder on two 28-core Xeon Platinum 8280 CPUs.

Tian, Jiannan↗

DOE OSTI2025

Sequence length scaling in vision transformers for scientific images on frontier

Vision Transformers (ViTs) are pivotal for foundational models in scientific imagery, including Earth science applications, due to their capability to process large sequence lengths. While transformers for text have inspired scaling sequence lengths in ViTs, adapting these for ViTs introduces unique challenges. We develop distributed sequence parallelism for ViTs, enabling them to handle up to 1M tokens. Our approach, leveraging DeepSpeed-Ulysses and Long-Sequence-Segmentation with model sharding, is the first to apply sequence parallelism in ViT training, achieving a 94% batch scaling efficiency on 2,048 AMD-MI250X GPUs. Evaluating sequence parallelism in ViTs, particularly in models up to 10B parameters, highlighted substantial bottlenecks. We countered these with hybrid sequence, pipeline, and flash attention strategies, to scale beyond single GPU memory limits. Our method significantly enhances climate modeling accuracy by 20% in temperature predictions, marking the first training of a vision transformer model to convergence with a sequence length of 188K tokens, using full self-attention.

Tsaris, Aristeidis (aris) [ORNL] (ORCID:0000000277↗

DOE OSTI2020

System with a direct contact desublimating heat exchanger and temperature regulation

A process to prevent fouling using a desublimating heat exchanger is disclosed. An outlet stream from the desublimating heat exchanger may be split into a plurality of parallel streams. The parallel streams may be sent through a plurality of discrete unit operations, and the unit operations may change the temperature of at least one of the parallel streams. Parallel streams of differing temperature may emerge from the unit operations. The parallel streams which are of a similar temperature may be mixed to form a warm stream and a cool stream. The warm stream and the cool stream may be sent to a mixing chamber. A mixed stream of substantially uniform temperature may emerge from the mixing chamber, and the mixed stream may be recycled back to the desublimating heat exchanger. The mixing chamber may be separate from the desublimating heat exchanger, or the parallel streams of differing temperature may be mixed in the desublimating heat exchanger.

42 ENGINEERING↗

DOE OSTI2021

Revisiting Huffman Coding: Toward Extreme Performance on Modern GPU Architectures

Today's high-performance computing (HPC) applications are producing vast volumes of data, which are challenging to store and transfer efficiently during the execution, such that data compression is becoming a critical technique to mitigate the storage burden and data movement cost. Huffman coding is arguably the most efficient Entropy coding algorithm in information theory, such that it could be found as a fundamental step in many modern compression algorithms such as DEFLATE. On the other hand, today's HPC applications are more and more relying on the accelerators such as GPU on supercomputers, while Huffman encoding suffers from low throughput on GPUs, resulting in a significant bottleneck in the entire data processing. In this paper, we propose and implement an efficient Huffman encoding approach based on modern GPU architectures, which addresses two key challenges: (1) how to parallelize the entire Huffman encoding algorithm, including codebook construction, and (2) how to fully utilize the high memory-bandwidth feature of modern GPU architectures. The detailed contribution is fourfold. (1) We develop an efficient parallel codebook construction on GPUs that scales effectively with the number of input symbols. (2) We propose a novel reduction based encoding scheme that can efficiently merge the codewords on GPUs. (3) We optimize the overall GPU performance by leveraging the state-of-the-art CUDA APIs such as Cooperative Groups. (4) We evaluate our Huffman encoder thoroughly using six real-world application datasets on two advanced GPUs and compare with our implemented multithreaded Huffman encoder. Experiments show that our solution can improve the encoding throughput by up to 5.0× and 6.8× on NVIDIA RTX 5000 and V100, respectively, over the state-of-the-art GPU Huffman encoder, and by up to 3.3× over the multithread encoder on two 28-core Xeon Platinum 8280 CPUs.

Tian, Jiannan↗

DOE OSTI2023

HEPnOS: a Specialized Data Service for High Energy Physics Analysis

In this paper, we present HEPnOS, a distributed data service for managing data produced by high-energy physics (HEP) experiments. Using HEPnOS, HEP applications can use HPC resources more effciently than traditional fle-based applications. The fle-based model leads to a rigid, chunk-based allocation of computational resources and limits the number of cores that can be used concurrently by an HEP application. The fundamental problem is that organizing domain-specifc data into fles inadvertently introduces a single, artifcial, confated tuning parameter that puts key optimization goals into confict: larger fle sizes reduce metadata overhead and thus improve I/O effciency, but smaller fle sizes provide more opportunity for workfow parallelism and load balancing. In this work, we introduce a domain-specifc data service that decouples that constraint so that data can be accessed and processed in its natural granularity while still maintaining I/O effciency. By removing the constraints introduced by fle handling we are able to obtain better scaling and make effcient use of more cores for processing a fxed-sized data sample. We demonstrate the improved scalability by using an application developed in the fle-based paradigm and comparing it to a version modifed to use HEPnOS.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2026

Biochemical parallels between catabolic pathways for lignin-associated aromatic dimers

Lignin is one of the most common biopolymers on Earth. In nature, lignin is primarily deconstructed by fungi into mixtures of aromatic compounds that are then assimilated by bacteria and fungi. Industrially, lignin is primarily generated as a byproduct of pulp and paper production and burned for process heat. However, if the appropriate assimilatory pathways were identified, deconstructed lignin could be funneled into value-added products using engineered bacteria. Foundational work has described pathways for assimilation of diverse monomeric aromatic compounds such as protocatechuate, ferulate, and syringate, as well as select dimers including those with β-O-4 and 5-5 interunit linkages. Recent advances have elucidated additional pathways for dimer assimilation, including pathways for new substrates as well as parallel pathways for previously characterized substrates. Comparing these dimer assimilation pathways can illuminate the underlying biochemical logic of assimilation for lignin-associated aromatic dimers and provide opportunities for metabolic engineering to enhance lignin valorization.

Sphingomonas↗

DOE OSTI2022

Stable parallel training of Wasserstein conditional generative adversarial neural networks

In this work, we propose a stable, parallel approach to train Wasserstein conditional generative adversarial neural networks (W-CGANs) under the constraint of a fixed computational budget. Differently from previous distributed GANs training techniques, our approach avoids inter-process communications, reduces the risk of mode collapse and enhances scalability by using multiple generators, each one of them concurrently trained on a single data label. The use of the Wasserstein metric also reduces the risk of cycling by stabilizing the training of each generator. We illustrate the approach on the CIFAR10, CIFAR100, and ImageNet1k datasets, three standard benchmark image datasets, maintaining the original resolution of the images for each dataset. Performance is assessed in terms of scalability and final accuracy within a limited fixed computational time and computational resources. To measure accuracy, we use the inception score, the Fréchet inception distance, and image quality. An improvement in inception score and Fréchet inception distance is shown in comparison to previous results obtained by performing the parallel approach on deep convolutional conditional generative adversarial neural networks as well as an improvement of image quality of the new images created by the GANs approach. Weak scaling is attained on both datasets using up to 2000 NVIDIA V100 GPUs on the OLCF supercomputer Summit.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

Effects of weave path parameters on the geometry of wire arc additive manufactured features

Here, the present work investigates the effect of process parameters on the geometry of wire arc additive manufactured parts. The geometric accuracy of features produced with a weaving strategy is compared to what can be accomplished with a typical overlapping bead strategy. In this work, single-layer and multi-layer geometries were deposited under varying process and path parameters. The wavelength, amplitude, and torch speed of the weaving path were varied, while the power, wire feed speed, and contact tip to work distance remained constant. The geometric deposition efficiencies of several samples produced with a weave strategy are directly compared to samples generated with two parallel overlapping beads with torch speeds defined to match the deposition rate of the weave samples. Feature geometries were characterized using optical microscopy and laser scanning data. The results indicate that implementing a weave path strategy can improve the geometric accuracy of wire arc additive features, thus increasing the effective volumetric deposition rate of the process. It is shown that the most consistent improvements resulted from the combination of low wavelength and high amplitude, which correlate to wider and taller printed layers.

GMAW↗