Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “gpu”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

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

At least 1,027 records · Page 57
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Energy Exascale Earth System Model v2.0.1

First patch release of v2.0.0 Changes since v2.0.0 [Important change] Fix ocean threading bug seen in debug cases on Chrysalis with Intel 20.0.4. Was introduced around time of v2.0.0 tag. Does not change v2.0.0 answers on Chrysalis because those didn't use threading or debugging. [EAM] Add semi-lagrangian tracer transport for theta-l (F90 and C++), add new algorithm for finding tropopause, add DSCREAM to allow v2 and SCREAM settings in same code such as adjust_ps [EAM-MMF] 60L default, allow C++ back end of RRTMGP (EAM too). [EAMxx] add nu-top functionality, fix forcing functor, add ttype9 and dcmip2012 tests 2.1, 2.2, and 3 HOMME: remove obsolete remap algs, option to specify dynamics alg indep of tracer, new sponge layer, add imex tests [ELM] Add topography-based subgrid (topounits), add FATES-ELM Nitro., Phos. and CH4 coupling, add land-use ts for NARRM, add lulc for SSP3 RCP7, Fix nutrient fertilization exp test and carbon isotope flux, Fix xactive lnd dry deposition, add lake water storage option, fix plant hydraulics 2d params, fix carbon budget calc, fix soil nutrient conc. bug, fix mosart dam bug, add test for new ELM, MOSART features, fix bug in O3 dry dep stomatal resistances, fix plant hydraulics restart BFB error, update mkmapdata. [MOSART] fix bug for reading the latitude from an unstructured input file, fix oversat in bubble test. [MPAS-ocean] Add CFC11, CFC12 tracers, add 2D spherical transport tests, fix del4 tracer mixing, add MARBL ocean tracer mixing, modify harmonic analysis options, add GPU port of vmix routines, fix calc of ML-averaged BV freq. [MPAS-seaice] Change extents of initial polar disks for oRRS18to6v3 grid, fix ice BGC with MARBL, update spherical test cases, fix DON coupling, Remove Cf from sea ice constants. [MPAS-landice] add CRYO1850-4xCO2 compset [CIME] add GCP, ANL GCE, Spock, Perlmutter, deprecate config_compilers.xml, fix and clean-up cmake macros, fix slurm bindings, refactor CIME internal testing, cleanup SCORPIO perf data, allow position independent compset naming, [also] update v2 benchmarking suite, extend e3sm_prod with throughput and memory checks

E3SM Project, DOE↗
DOE OSTI2021

Efficient exascale discretizations: High-order finite element methods

Efficient exploitation of exascale architectures requires rethinking of the numerical algorithms used in many large-scale applications. These architectures favor algorithms that expose ultra fine-grain parallelism and maximize the ratio of floating point operations to energy intensive data movement. One of the few viable approaches to achieve high efficiency in the area of PDE discretizations on unstructured grids is to use matrix-free/partially assembled high-order finite element methods, since these methods can increase the accuracy and/or lower the computational time due to reduced data motion. In this paper we provide an overview of the research and development activities in the Center for Efficient Exascale Discretizations (CEED), a co-design center in the Exascale Computing Project that is focused on the development of next-generation discretization software and algorithms to enable a wide range of finite element applications to run efficiently on future hardware. CEED is a research partnership involving more than 30 computational scientists from two US national labs and five universities, including members of the Nek5000, MFEM, MAGMA and PETSc projects. We discuss the CEED co-design activities based on targeted benchmarks, miniapps and discretization libraries and our work on performance optimizations for large-scale GPU architectures. We also provide a broad overview of research and development activities in areas such as unstructured adaptive mesh refinement algorithms, matrix-free linear solvers, high-order data visualization, and list examples of collaborations with several ECP and external applications.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

EXAGRAPH: Graph and combinatorial methods for enabling exascale applications

Combinatorial algorithms in general and graph algorithms in particular play a critical enabling role in numerous scientific applications. However, the irregular memory access nature of these algorithms makes them one of the hardest algorithmic kernels to implement on parallel systems. With tens of billions of hardware threads and deep memory hierarchies, the exascale computing systems in particular pose extreme challenges in scaling graph algorithms. The codesign center on combinatorial algorithms, ExaGraph, was established to design and develop methods and techniques for efficient implementation of key combinatorial (graph) algorithms chosen from a diverse set of exascale applications. Algebraic and combinatorial methods have a complementary role in the advancement of computational science and engineering, including playing an enabling role on each other. In this paper, we survey the algorithmic and software development activities performed under the auspices of ExaGraph from both a combinatorial and an algebraic perspective. In particular, we detail our recent efforts in porting the algorithms to manycore accelerator (GPU) architectures. We also provide a brief survey of the applications that have benefited from the scalable implementations of different combinatorial algorithms to enable scientific discovery at scale. We believe that several applications will benefit from the algorithmic and software tools developed by the ExaGraph team.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Compressed basis GMRES on high-performance graphics processing units

Krylov methods provide a fast and highly parallel numerical tool for the iterative solution of many large-scale sparse linear systems. To a large extent, the performance of practical realizations of these methods is constrained by the communication bandwidth in current computer architectures, motivating the investigation of sophisticated techniques to avoid, reduce, and/or hide the message-passing costs (in distributed platforms) and the memory accesses (in all architectures). This article leverages Ginkgo’s memory accessor in order to integrate a communication-reduction strategy into the (Krylov) GMRES solver that decouples the storage format (i.e., the data representation in memory) of the orthogonal basis from the arithmetic precision that is employed during the operations with that basis. Given that the execution time of the GMRES solver is largely determined by the memory accesses, the cost of the datatype transforms can be mostly hidden, resulting in the acceleration of the iterative step via a decrease in the volume of bits being retrieved from memory. Together with the special properties of the orthonormal basis (whose elements are all bounded by 1), this paves the road toward the aggressive customization of the storage format, which includes some floating-point as well as fixed-point formats with mild impact on the convergence of the iterative process. We develop a high-performance implementation of the “compressed basis GMRES” solver in the Ginkgo sparse linear algebra library using a large set of test problems from the SuiteSparse Matrix Collection. We demonstrate robustness and performance advantages on a modern NVIDIA V100 graphics processing unit (GPU) of up to 50% over the standard GMRES solver that stores all data in IEEE double-precision.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Orchestration of materials science workflows for heterogeneous resources at large scale

In the era of big data, materials science workflows need to handle large-scale data distribution, storage, and computation. Any of these areas can become a performance bottleneck. We present a framework for analyzing internal material structures (e.g., cracks) to mitigate these bottlenecks. We demonstrate the effectiveness of our framework for a workflow performing synchrotron X-ray computed tomography reconstruction and segmentation of a silica-based structure. Our framework provides a cloud-based, cutting-edge solution to challenges such as growing intermediate and output data and heavy resource demands during image reconstruction and segmentation. Specifically, our framework efficiently manages data storage, scaling up compute resources on the cloud. The multi-layer software structure of our framework includes three layers. A top layer uses Jupyter notebooks and serves as the user interface. A middle layer uses Ansible for resource deployment and managing the execution environment. A low layer is dedicated to resource management and provides resource management and job scheduling on heterogeneous nodes (i.e., GPU and CPU). At the core of this layer, Kubernetes supports resource management, and Dask enables large-scale job scheduling for heterogeneous resources. The broader impact of our work is four-fold: through our framework, we hide the complexity of the cloud’s software stack to the user who otherwise is required to have expertise in cloud technologies; we manage job scheduling efficiently and in a scalable manner; we enable resource elasticity and workflow orchestration at a large scale; and we facilitate moving the study of nonporous structures, which has wide applications in engineering and scientific fields, to the cloud. While we demonstrate the capability of our framework for a specific materials science application, it can be adapted for other applications and domains because of its modular, multi-layer architecture.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

GenSLMs: Genome-scale language models reveal SARS-CoV-2 evolutionary dynamics

We seek to transform how new and emergent variants of pandemic-causing viruses, specifically SARS-CoV-2, are identified and classified. By adapting large language models (LLMs) for genomic data, we build genome-scale language models (GenSLMs) which can learn the evolutionary landscape of SARS-CoV-2 genomes. By pre-training on over 110 million prokaryotic gene sequences and fine-tuning a SARS-CoV-2-specific model on 1.5 million genomes, we show that GenSLMs can accurately and rapidly identify variants of concern. Thus, to our knowledge, GenSLMs represents one of the first whole-genome scale foundation models which can generalize to other prediction tasks. We demonstrate scaling of GenSLMs on GPU-based supercomputers and AI-hardware accelerators utilizing 1.63 Zettaflops in training runs with a sustained performance of 121 PFLOPS in mixed precision and peak of 850 PFLOPS. We present initial scientific insights from examining GenSLMs in tracking evolutionary dynamics of SARS-CoV-2, paving the path to realizing this on large biological data.

Zvyagin, Maxim↗
DOE OSTI2024

AMReX and pyAMReX: Looking beyond the exascale computing project

AMReX is a software framework for the development of block-structured mesh applications with adaptive mesh refinement (AMR). AMReX was initially developed and supported by the AMReX Co-Design Center as part of the U.S. DOE Exascale Computing Project (ECP), and is continuing to grow post-ECP. In addition to adding new functionality and performance improvements to the core AMReX framework, we have also developed a Python binding, pyAMReX, that provides a bridge between AMReX-based application codes and the data science ecosystem. pyAMReX provides zero-copy application GPU data access for AI/ML, in situ analysis and application coupling, and enables rapid, massively parallel prototyping. In this paper we review the overall functionality of AMReX and pyAMReX, focusing on new developments, new functionality, and optimizations of key operations. We also summarize capabilities of ECP projects that used AMReX and provide an overview of new, non-ECP applications.

Myers, Andrew↗
DOE OSTI2024

The ECP SICM project: Managing complex memory hierarchies for exascale applications

The Exascale Computing Project (ECP)’s Simplified Interface to Complex Memories (SICM) effort focuses on developing universal interfaces for discovering, managing, and sharing data across complex memory hierarchies. These facilitate the exploitation of emerging memory technologies and support precise control over their various trade-offs such as high-bandwidth versus low-latency, persistent versus ephemeral, high-capacity versus low-capacity, and near-CPU versus near-GPU. SICM comprises three interrelated components: a low-level interface, a high-level interface, and a persistent-heap interface. The low-level SICM interface is intended for system and run-time developers as well as expert application developers who prefer full control of the memory objects used within their application. The high-level SICM interface builds upon the low-level interface, employing application-level profiling and analysis to optimize data management for complex memory hierarchies. The persistent-heap interface provides applications with a persistent memory allocator that can allocate custom C++ data structures in both block-storage and byte-addressable persistent memories.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Asynchronous-many-task systems: Challenges and opportunities - Scaling an AMR astrophysics code on exascale machines using Kokkos and HPX

Dynamic and adaptive mesh refinement is pivotal in high-resolution, multi-physics, multi-model simulations, necessitating precise physics resolution in localized areas across expansive domains. Today’s supercomputers’ extreme heterogeneity presents a significant challenge for dynamically adaptive codes, highlighting the importance of achieving performance portability at scale. Our research focuses on astrophysical simulations, particularly stellar mergers, to elucidate early universe dynamics. Here, we present Octo-Tiger, leveraging Kokkos, HPX, and SIMD for portable performance at scale in complex, massively parallel adaptive multi-physics simulations. Octo-Tiger supports diverse processors, accelerators, and network backends. Experiments demonstrate exceptional scalability across several heterogeneous supercomputers including Perlmutter, Frontier, and Fugaku, encompassing major GPU architectures and x86, ARM, and RISC-V CPUs. Parallel efficiency of 47.59% (110,080 cores and 6880 hybrid A100 GPUs) on a full-system run on Perlmutter (26% HPCG peak performance) and 51.37% (using 32,768 cores and 2048 MI250X) on Frontier are achieved.

97 MATHEMATICS AND COMPUTING↗
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 OSTI2026

ORCHA: A performance portability system for extreme heterogeneity

Heterogeneity is the prevalent trend in the rapidly evolving high-performance computing (HPC) landscape in both hardware and application software. The diversity in hardware platforms, currently comprising various accelerators and a future possibility of specializable chiplets, poses a significant challenge for scientific software developers aiming to harness optimal performance across different computing platforms while maintaining the quality of solutions when their applications are simultaneously growing more complex. Code synthesis and code generation can provide mechanisms to mitigate this challenge. We have developed a divide and conquer approach where different aspects of performance are handled by different stand-alone tools that are interfaced with the application through generated code. This portability system, ORCHA, enables users to configure and orchestrate their computations among available resources on a platform by specifying a high-level recipe, thereby permitting a many-to-many paradigm where each recipe results in a different variant of the application. The core design goal is to let users decide the application’s hardware mapping and orchestration by editing only the high-level recipe—without modifying the maintained source code or binding the application to a particular runtime system. Tools in ORCHA distribution are: CG-Kit for translating the recipe into an execution graph; Milhoja to execute the graph by orchestrating data and task movement among hardware resources; and Macroprocessor that enables users to define their own code-shorthand for higher composability and easier management of code variants. Additionally, the design of ORCHA permits tools to work in a plug-and-play mode where the application can build and run without CG-Kit and Milhoja, and either tool can be swapped out for other tools with similar capabilities by modifying the code generation portion of ORCHA. In this paper, we describe the design of ORCHA and the role that code-generation plays in isolating applications from tools. We demonstrate the breadth of configurations ORCHA enables with a case study in which an application configuration is realized on three distinct hardware mappings—a GPU-centric, a CPU/GPU balanced, and a CPU/GPU concurrent layouts by using different recipes.

Lee, Youngjun↗
DOE OSTI2025

An open-access simulated earthquake ground-motion database for an M7 Hayward Fault earthquake in the San Francisco Bay Region

Comprehensive understanding of earthquake ground motions, particularly in the near-fault region of large-magnitude events, is limited by gaps in strong-motion data. This challenge is prominent in areas with high seismic hazard but infrequent large earthquakes where data is sparse and difficult to interpret. These data limitations lead to uncertainties in the development of site-specific ground motions, which are crucial for engineering risk assessments. To address these challenges, physics-based regional-scale ground-motion simulations have been developed. With the emergence of exaflop-scale computing ecosystems, it is now possible to simulate regional earthquake processes at unprecedented fidelity and generate the large number of fault rupture realizations necessary to characterize both intra- and inter-event ground-motion variability. This article introduces a new database of simulated earthquake ground motions, created for applications in earthquake engineering, earthquake planning, and emergency response. The inaugural version of the database features simulated ground motions for a magnitude 7 Hayward Fault earthquake in the San Francisco Bay Region (SFBR), using the EarthQuake SIMulation (EQSIM) simulation framework and the Graves–Pitarka kinematic rupture model. The aim is to provide high-fidelity, spatially dense, three-component motions generated on the Department of Energy’s (DOE) newest generation of graphics processing unit (GPU)-accelerated supercomputers. These motions are being made openly available to the engineering, scientific, and disaster planning communities. In addition, this work develops protocols for the efficient dissemination of these large data sets and emphasizes community engagement to build confidence in their application. This article discusses the methodology behind the data, underlying software verification and validation, scalable data management, and a user interface for data access. The goal is to facilitate widespread use and elicit expert feedback to maximize the utility and exploitation of simulated motions. While the initial focus is on the San Francisco Region, simulations for additional regions will be added as the DOE program progresses.

Simulated ground-motion database↗
DOE OSTI2024

Road Lidar Dataset for the TxDOT Austin District

This is a road lidar data collection for developing road elevation models and road inundation mapping methodologies, a joint work between ORNL and The University of Texas at Austin. This dataset is generated as part of the flood transportation infrastructure, partly funded by the NOAA CIROH project. ORNL is a project partner for high-performance computing-empowered flood inundation mapping methodology R&D. The dataset is computed using a GPU-accelerated lidar data processing workflow developed at ORNL. The lidar data source is from TxGIO, the state lidar data collection site. The output dataset is in two formats: laz and copc. It is organized by TxDOT's maintenance sections, covering the Austin District. Data size: 3.86 billion road lidar points, 1.67% of the entire lidar data input Projection: EPSG:32614 (WGS84/UTM zone 14N) Website: https://web.corral.tacc.utexas.edu/nfiedata/road3d/austin_district/AustinMaintenanceSections_H_epsg6343_V_epsg5703/ LICENSE FOR USE -- MAPS AND DATA DISCLAIMER This resource is shared under the Creative Commons Attribution CC BY, http://creativecommons.org/licenses/by/4.0/ MAPS AND DATA DISCLAIMER The Oak Ridge National Laboratory (ORNL) shall not be held liable for improper or incorrect use of the data described or information contained on this map or associated series of maps. The data and related map graphics are not legal, land survey or engineering documents and are not intended to be used as such. ORNL gives no warranty, express or implied, as to the accuracy, reliability, utility or completeness of this information. The user of these maps and data assumes all responsibility and risk for the use of the maps and data. ORNL disclaims all warranties, representations or endorsements either express or implied, with regard to the information contained in this map product, including, but not limited to, all implied warranties of merchantability, fitness for a particular purpose or non-infringement. This preliminary map product is for research and review purposes only. It is not intended to be used for emergency management operational or life safety decisions at the local or regional governmental level or by the general public. Users requiring information regarding hazardous conditions or meteorological conditions for specific geographic areas should consult directly with their city or county emergency management office.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2026

Dataset for Top Model Decision Tree: Selecting Segmentation Models for Reliable Quantitative Analysis in Low- and Ultralow-Dose CryoEM

Motivation Multiple deep learning model architectures can be used to segment bacterial membranes in cryoEM images. However, an AI-based tool advancement is often presented with only a single segmentation model for broad use, and this single model may show inconsistent results across datasets from different users. Here, we present the Top Model Decision Tree, a model screening framework to screen for the best model to generate bacterial inner and outer membrane masks based on user priorities. We use pre-trained segmentation models from YOLOv11, YOLO26, U-Net, Detectron2 and SAM3 fine-tuned on bacterial inner and outer membranes imaged with cryoEM. Run the Framework This notebook must be opened in Google Colab. Mount Google Drive and run with a GPU-based runtime. Open the notebook and follow steps to git clone in folders and files within this repository. There will be a repeating top_model_decision_tree.ipynb (notebook clone) that will not be used. Save your .png binary mask files and .csv table outputs within your Google Drive or download before closing the notebook. The models and all analysis/training scripts are available at [GitHub: https://github.com/Lynnicia/CryoEM_membranes_top_model_decision_tree and https://github.com/Sireesiru/Semantic-Segmentation-of-bacterial-cell-envelope-using-U-Nets.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2022

High-performance and high-fidelity Monte Carlo solutions to the BEAVRS benchmark

The BEAVRS (Benchmark for Evaluation and Validation of Reactor Simulation) benchmark is solved by PRAGMA, the GPU-based continuous energy Monte Carlo code. The resulting solutions are comprised of the detailed simulation results of two cycles, each of which consists of the zero power physics test (ZPPT) and the core depletion calculations. The ZPPT consists of characteristic parameters, such as critical boron concentration (CBC), control rod bank worth, isothermal temperature coefficients, and assembly-wise detector signal, which are compared with measured data provided by the benchmark administration. The core depletion calculations were performed for both the hot full power and the load follow modes, and the comparison was made with the measured or deduced CBCs and assembly-wise detector signals. In the load follow calculations, the operating power history was approximately applied to simulate the real operation as closely as possible. PRAGMA performed the various calculations with a tremendous number of histories ranging up to hundreds of millions per cycle, exploiting GPUs' massively parallel performance. The load follow run time was shorter than 16 hours on a single rack of computing nodes mounded with 24 gaming GPUs with a remarkable agreement with the measurements for most comparisons. (authors)

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2024

Flexible silicon photonic architecture for accelerating distributed deep learning

The increasing size and complexity of deep learning (DL) models have led to the wide adoption of distributed training methods in datacenters (DCs) and high-performance computing (HPC) systems. However, communication among distributed computing units (CUs) has emerged as a major bottleneck in the training process. In this study, we propose Flex-SiPAC, a flexible silicon photonic accelerated compute cluster designed to accelerate multi-tenant distributed DL training workloads. Flex-SiPAC takes a co-design approach that combines a silicon photonic hardware platform with a tailored collective algorithm, optimized to leverage the unique physical properties of the architecture. The hardware platform integrates a novel wavelength-reconfigurable transceiver design and a micro-resonator-based wavelength-reconfigurable switch, enabling the system to achieve flexible bandwidth steering in the wavelength domain. The collective algorithm is designed to support reconfigurable topologies, enabling efficient all-reduce communications that are commonly used in DL training. The feasibility of the Flex-SiPAC architecture is demonstrated through two testbed experiments. First, an optical testbed experiment demonstrates the flexible routing of wavelengths by shuffling an array of input wavelengths using a custom-designed spatial-wavelength selective switch. Second, a four-GPU testbed running two DL workloads shows a 23% improvement in job completion time compared to a similarly sized leaf-spine topology. We further evaluate Flex-SiPAC using large-scale simulations, which show that Flex-SiPAC is able to reduce the communication time by 26% to 29% compared to state-of-the-art compute clusters under representative collective operations.

Wu, Zhenguo (ORCID:0000000322847985)↗
DOE OSTI2021

OpenAWSEM with Open3SPN2: A fast, flexible, and accessible framework for large-scale coarse-grained biomolecular simulations

We present OpenAWSEM and Open3SPN2, new cross-compatible implementations of coarse-grained models for protein (AWSEM) and DNA (3SPN2) molecular dynamics simulations within the OpenMM framework. These new implementations retain the chemical accuracy and intrinsic efficiency of the original models while adding GPU acceleration and the ease of forcefield modification provided by OpenMM’s Custom Forces software framework. By utilizing GPUs, we achieve around a 30-fold speedup in protein and protein-DNA simulations over the existing LAMMPS-based implementations running on a single CPU core. We showcase the benefits of OpenMM’s Custom Forces framework by devising and implementing two new potentials that allow us to address important aspects of protein folding and structure prediction and by testing the ability of the combined OpenAWSEM and Open3SPN2 to model protein-DNA binding. The first potential is used to describe the changes in effective interactions that occur as a protein becomes partially buried in a membrane. We also introduced an interaction to describe proteins with multiple disulfide bonds. Using simple pairwise disulfide bonding terms results in unphysical clustering of cysteine residues, posing a problem when simulating the folding of proteins with many cysteines. We now can computationally reproduce Anfinsen’s early Nobel prize winning experiments by using OpenMM’s Custom Forces framework to introduce a multi-body disulfide bonding term that prevents unphysical clustering. Our protein-DNA simulations show that the binding landscape is funneled towards structures that are quite similar to those found using experiments. In summary, this paper provides a simulation tool for the molecular biophysics community that is both easy to use and sufficiently efficient to simulate large proteins and large protein-DNA systems that are central to many cellular processes. These codes should facilitate the interplay between molecular simulations and cellular studies, which have been hampered by the large mismatch between the time and length scales accessible to molecular simulations and those relevant to cell biology.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2020

Traversing large graphs on GPUs with unified memory

Due to the limited capacity of GPU memory, the majority of prior work on graph applications on GPUs has been restricted to graphs of modest sizes that fit in memory. Recent hardware and software advances make it possible to address much larger host memory transparently as a part of a feature known as unified virtual memory. While accessing host memory over an interconnect is understandably slower, the problem space has not been sufficiently explored in the context of a challenging workload with low computational intensity and an irregular data access pattern such as graph traversal. We analyse the performance of breadth first search (BFS) for several large graphs in the context of unified memory and identify the key factors that contribute to slowdowns. Next, we propose a lightweight offline graph reordering algorithm, HALO (Harmonic Locality Ordering), that can be used as a pre-processing step for static graphs. HALO yields speedups of 1.5x-1.9x over baseline in subsequent traversals. Our method specifically aims to cover large directed real world graphs in addition to undirected graphs whereas prior methods only account for the latter. Additionally, we demonstrate ties between the locality ordering problem and graph compression and show that prior methods from graph compression such as recursive graph bisection can be suitably adapted to this problem.

Gera, Prasun↗