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 145 records · Page 8
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

Performance Profile of Transformer Fine-Tuning in Multi-GPU Cloud Environments

The study presented here focuses on performance characteristics and trade-offs associated with running machine-learning tasks in multi-GPU environments on both on-site cloud computing resources and commercial cloud services (Azure). Specifically, this study examines these tradeoffs by examining the performance of training and fine-tuning of transformer-based deep-learning (DL) networks on clinical notes and data, a task of critical importance in the medical domain. To this end, we perform DL-related experiments on the widely deployed NVIDIA V100 GPUs and on the newer A100 GPUs connected via NVLink or PCIe. This study analyzes the execution time of major operations to train DL models and investigate popular options to optimize each of them. We examine and present the findings on the impacts that various operations (e.g. data loading into GPUs, training, fine-tuning), optimizations, and system configurations (single vs. multi-GPU, NVLink vs. PCIe) have on the overall training performance.

Begoli, Edmon↗
DOE OSTI2020

LOGAN: High-Performance GPU-Based X-Drop Long-Read Alignment

Pairwise sequence alignment is one of the most computationally intensive kernels in genomic data analysis, accounting for more than 90% of the runtime for key bioinformatics applications. This method is particularly expensive for third-generation sequences due to the high computational cost of analyzing sequences of length between 1Kb and 1Mb. Given the quadratic overhead of exact pairwise algorithms for long alignments, the community primarily relies on approximate algorithms that search only for high-quality alignments and stop early when one is not found. In this work, we present the first GPU optimization of the popular X-drop alignment algorithm, that we named LOGAN. Results show that our high-performance multi-GPU implementation achieves up to 181.6 GCUPS and speed-ups up to 6.6× and 30.7× using 1 and 6 NVIDIA Tesla V100, respectively, over the state-of-the-art software running on two IBM Power9 processors using 168 CPU threads, with equivalent accuracy. We also demonstrate a 2.3× LOGAN speed-up versus ksw2, a state-of-art vectorized algorithm for sequence alignment implemented in minimap2, a long-read mapping software. Furthermore, to highlight the impact of our work on a real-world application, we couple LOGAN with a many-to-many long-read alignment software called BELLA, and demonstrate that our implementation improves the overall BELLA runtime by up to 10.6×. Finally, we adapt the Roofline model for LOGAN and demonstrate that our implementation is near optimal on the NVIDIA Tesla V100s.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

MICCO: An Enhanced Multi-GPU Scheduling Framework for Many-Body Correlation Functions

Calculation of many-body correlation functions is one of the critical kernels utilized in many scientific computing areas, especially in Lattice Quantum Chromodynamics (Lattice QCD). It is formalized as a sum of a large number of contraction terms each of which can be represented by a graph consisting of vertices describing quarks inside a hadron node and edges designating quark propagations at specific time intervals. Due to its computation- and memory-intensive nature, real-world physics systems (e.g., multi-meson or multi-baryon systems) explored by Lattice QCD prefer to leverage multi-GPUs. Different from general graph processing, many-body correlation function calculations show two specific features: a large number of computation-/data-intensive kernels and frequently repeated appearances of original and intermediate data. The former results in expensive memory operations such as tensor movements and evictions. The latter offers data reuse opportunities to mitigate the data-intensive nature of many-body correlation function calculations. However, existing graph-based multi-GPU schedulers cannot capture these data-centric features, thus resulting in a sub-optimal performance for many-body correlation function calculations. To address this issue, this paper presents a multi-GPU scheduling framework, MICCO, to accelerate contractions for correlation functions particularly by taking the data dimension (e.g., data reuse and data eviction) into account. This work first performs a comprehensive study on the interplay of data reuse and load balance, and designs two new concepts: local reuse pattern and reuse bound to study the opportunity of achieving the optimal trade-off between them. Based on this study, MICCO proposes a heuristic scheduling algorithm and a machine-learning-based regression model to generate the optimal setting of reuse bounds. Specifically, MICCO is integrated into a real-world Lattice QCD system, Redstar, for the first time running on multiple GPUs. The evaluation demonstrates MICCO outperforms other state-of-art works, achieving up to 2.25× speedup in synthesized datasets, and 1.49× speedup in real-world correlation functions.

Wang, Qihan↗
DOE OSTI2022

Integer Sum Reduction with OpenMP on an AMD MI100 GPU

Sum reduction is a primitive operation in parallel computing. Device offload support allows a user to use OpenMP directives to take advantage of a highly capable GPU. In this paper, we present the integer sum reduction annotated with the OpenMP directives and evaluate the performance impacts of tunable parameters with the AOMP and GCC compilers on an AMD MI100 GPU. In addition, we explain the implementations of the OpenMP reduction by the compilers. Sweeping over the pruned parameter space, we find that the speedup is approximately 20 with AOMP, and the reduction performance using AOMP is approximately 11% higher than that using GCC. However, the OpenMP offload performance is approximately 30% lower compared to the performance of the reductions written with rocThrust or hipCUB.

Jin, Zheming↗
DOE OSTI2025

Performance of an optical TPC Geant4 simulation with opticks GPU-accelerated photon propagation

We investigate the performance of Opticks, a NVIDIA OptiX API 7.5 GPU-accelerated photon propagation tool compared with a single-threaded Geant4 simulation. We compare the simulations using an improved model of the NEXT-CRAB-0 gaseous time projection chamber. Performance results suggest that Opticks improves simulation speeds by between 58.47 ± 0.02 and 181.39 ± 0.28 times relative to a CPU-only Geant4 simulation and these results vary between different types of GPU and CPU. A detailed comparison shows that the number of detected photons, along with their times and wavelengths, are in good agreement between Opticks and Geant4.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2023

Application Experiences on a GPU-Accelerated Arm-based HPC Testbed

This paper assesses and reports the experience of ten teams working to port, validate, and benchmark several High Performance Computing applications on a novel GPU-accelerated Arm testbed system. The testbed consists of eight NVIDIA Arm HPC Developer Kit systems, each one equipped with a server-class Arm CPU from Ampere Computing and two data center GPUs from NVIDIA Corp. The systems are connected together using InfiniBand interconnect. The selected applications and mini-apps are written using several programming languages and use multiple accelerator-based programming models for GPUs such as CUDA, OpenACC, and OpenMP offloading. Working on application porting requires a robust and easy-to-access programming environment, including a variety of compilers and optimized scientific libraries. The goal of this work is to evaluate platform readiness and assess the effort required from developers to deploy well-established scientific workloads on current and future generation Arm-based GPU-accelerated HPC systems. The reported case studies demonstrate that the current level of maturity and diversity of software and tools is already adequate for large-scale production deployments.

Elwasif, Wael↗
DOE OSTI2023

A GPU-Accelerated Population Generation, Sorting, and Mutation Kernel for an Optimization-Based Causal Inference Model

We develop a GPU-accelerated machine learning generative adversarial network model that can be used with observational data for the purpose of constructing causal inferences. The theoretical basis of our machine learning model is novel and is conceptualized to be operable and scalable for high performance computing platforms. Our GPU-accelerated code enables large-scale parallelization of the computation within a common and accessible computing environment. This will expand the reach of our model and empower research in new substantive domains while maintaining the underlying theoretical properties.

Cho, Wendy K. Tam↗
DOE OSTI2023

Performance Evaluation of Heterogeneous GPU Programming Frameworks for Hemodynamic Simulations

Preparing for the deployment of large scientific and engineering codes on upcoming exascale systems with GPU-dense nodes is made challenging by the unprecedented diversity of device architectures and heterogeneous programming models. In this work, we evaluate the process of porting a massively parallel, fluid dynamics code written in CUDA to SYCL, HIP, and Kokkos with a range of backends, using a combination of automated tools and manual tuning. We use a proxy application along with a custom performance model to inform the results and identify additional optimization strategies. At scale performance of the programming model implementations are evaluated on pre-production GPU node architectures for Frontier and Aurora, as well as on current NVIDIA device-based systems Summit and Polaris. Real-world workloads representing 3D blood flow calculations in complex vasculature are assessed. Our analysis highlights critical trade-offs between code performance, portability, and development time.

Martin, Aristotle↗
DOE OSTI2023

cuAlign: Scalable Network Alignment on GPU Accelerators

Given two graphs, the objective of network alignment is to find the best one-to-one mapping of vertices in one graph (??) to vertices in the other (??), such that the number of overlaps is maximized. We say that edges(??, ??) ???and(??', ??') ??? are overlapped if ?? is mapped to ??' and ?? is mapped to??'. Network alignment is an important optimization problem with several applications in bioinformatics, computer vision and ontology matching. Since it is an NP-hard problem, efficient heuristics and scalable implementations are necessary. In this work, we introduce a new framework that combines the concepts of intra-network proximity using vertex embedding,Belief Propagation (BP) and approximate weighted matching, and provides qualitative improvements up to22%over state-of-the-art approaches. We also provide scalable implementations on GPU accelerators, demonstrating up to19×speedup for Belief Propagation and 3× speedup for approximate weighted matching relative to previous multithreaded implementation. A combination of combinatorial and algebraic kernels within the network alignment algorithm poses significant hurdles for parallelization. Load imbalance and irregular DRAM traffic limit achievable performance on GPUs. Our novel approach identifies and exploits unique structural proper-ties of the BP-based algorithm and employs code fusion to reduce data movement between different steps of the algorithm. Using a diverse set of inputs, we demonstrate qualitative improvements of our algorithms, and performance gains of our GPU-accelerated implementation. We believe that our work will enable algorithmic improvements and practical applications of network alignment.

Xiang, Lizhi↗
DOE OSTI2024

Understanding GPU Memory Corruption at Extreme Scale: The Summit Case Study

GPU memory corruption and in particular double-bit errors (DBEs) remain one of the least understood aspects of HPC system reliability. Albeit rare, their occurrences always lead to job termination and can potentially cost thousands of node-hours, either from wasted computations or as the overhead from regular checkpointing needed to minimize the losses. As supercomputers and their components simultaneously grow in scale, density, failure rates, and environmental footprint, the efficiency of HPC operations becomes both an imperative and a challenge. We examine DBEs using system telemetry data and logs collected from the Summit supercomputer, equipped with 27,648 Tesla V100 GPUs with 2nd-generation high-bandwidth memory (HBM2). Using exploratory data analysis and statistical learning, we extract several insights about memory reliability in such GPUs. We find that GPUs with prior DBE occurrences are prone to experience them again due to otherwise harmless factors, correlate this phenomenon with GPU placement, and suggest manufacturing variability as a factor. On the general population of GPUs, we link DBEs to short- and long-term high power consumption modes while finding no significant correlation with higher temperatures. We also show that the workload type can be a factor in memory’s propensity to corruption.

Oles, Vlad↗
DOE OSTI2025

TrioSim: A Lightweight Simulator for Large-Scale DNN Workloads on Multi-GPU Systems

Deep Neural Networks (DNNs) have become increasingly capable of performing tasks ranging from image recognition to content generation. The training and inference of DNNs heavily rely on GPUs, as GPUs' massively parallel architecture delivers extremely high computing capability. With the growing complexity of DNNs and the size of training datasets, training DNNs with a large number of GPUs is becoming a prevalent strategy. Researchers have been exploring how to design software and hardware systems for GPU farms to achieve the best utilization, efficiency, and DNN accuracy during training or inference. However, when designing and deploying such systems, designers usually rely on testing on physical hardware platforms equipped with many GPUs, incurring high costs that are almost prohibitive for system designers to test different configurations and designs, even for highly resourceful companies. While an alternative solution is to test on GPU simulators, they are often too slow for these l

Li, Ying [William & Mary, Williamsburg, VA, USA] (↗
DOE OSTI2025

Performance Impact and Trade-Offs for Tuning Key Architectural Parameters on CPU+GPU Systems

In this work, we performed an initial design space exploration of an accelerated processing unit (APU)—a hybrid CPU+GPU architecture that integrates both compute units (CUs) and memory into a unified system. This integration aims to reduce data movement, enhance memory locality, and improve energy efficiency by enabling the CPU and GPU to share memory directly. This effort focused on the interplay of key design components—cache line size, the number of CUs, and main memory technology—and the trade-offs of each configuration were analyzed. This paper highlights the various configurations’ impact on memory accesses, data reuse, and power utilization. The results provide valuable insights that can be leveraged to optimize APU architectures for high-performance and energy-efficient computing and thus create a balanced architecture. This optimization can be achieved by adopting dynamic cache management, runtime CU scaling, and advanced memory integration, highlighting the potential of APUs to address critical challenges in compute, data movement, and memory power consumption.

Asifuzzaman, Kazi [ORNL] (ORCID:0000000240044791)↗
DOE OSTI2025

Characterizing the Impact of GPU Power Management on an Exascale System

As GPU-accelerated high-performance computing (HPC) systems approach exascale performance, controlling energy consumption without compromising throughput is essential. Architectures such as the AMD MI250X-based Frontier supercomputer provide runtime mechanisms like frequency and power capping, enabling energy tuning without modifying application code. Although both target energy reduction, they operate via distinct hardware control paths and influence workloads differently. We present a comprehensive evaluation of these strategies on a leadership-class system using diverse HPC proxy applications representative of production workloads. Our study analyzes performance–energy trade-offs across multiple capping levels, node counts (1 and 32), and application profiles. Results show that frequency capping generally achieves higher energy efficiency and scalability, with gains of up to 13.2% without performance loss, while power capping is more effective for single-node runs or bursty GPU utilization. We also provide practical guidelines to help system administrators and users balance energy efficiency and performance in large-scale scientific workloads.

Costa, Mariana [Universidade Federal do Rio Grande↗
DOE OSTI2025

TorchBraid: High-Performance Layer-Parallel Training of Deep Neural Networks with MPI and GPU Acceleration

TorchBraid is a high-performance implementation of layer-parallel training for deep neural networks (DNNs) supporting MPI-based parallelism and GPU acceleration. Layer-parallel training has been developed to overcome the serialization inherent in forward and backward propagation of DNNs that limits utilization of computational resources in the strong scaling limit. To achieve this, TorchBraid integrates the PyTorch neural network framework with the state-of-the-art XBraid time-parallel library. Furthermore, this article presents the use and performance of TorchBraid, in addition to solutions for overcoming the algorithmic challenges inherent in combining automatic differentiation with layer-parallel. Results are presented with and without GPU acceleration for the Tiny ImageNet and MNIST image classification data sets, as well as recurrent neural networks. Overall, TorchBraid enables fast training of DNNs, both in a strong and weak scaling context. In addition to the TorchBraid software, several new advances in applying layer-parallel algorithms are detailed. Integration of layer-parallel with data-parallel algorithms is presented for the first time, showing the computational advantages of the combination. Standard deep learning techniques, like batch-normalization, are developed for layer-parallel training. Finally, a new approach combining layer-parallel with spatial coarsening in order to accelerate training for 3D image classification shows roughly a 10× speedup over serial execution.

Layer-parallel↗
DOE OSTI2022

MemLeap GPU Memory Profiler

MemLeap is a GPU Memory Profiler built atop Nvbit (NVidia Binary Instrumentation Tool) framework. It analyzes memory accesses to memory objects in GPU memory, including Unified Memory, to optimize memory usage and reveal performance bottlenecks.

Peng, Bo↗
DOE OSTI2023

End-to-end GPU acceleration of low-order-refined preconditioning for high-order finite element discretizations

In this article, we present algorithms and implementations for the end-to-end GPU acceleration of matrix-free low-order-refined preconditioning of high-order finite element problems. The methods described here allow for the construction of effective preconditioners for high-order problems with optimal memory usage and computational complexity. The preconditioners are based on the construction of a spectrally equivalent low-order discretization on a refined mesh, which is then amenable to, for example, algebraic multigrid preconditioning. The constants of equivalence are independent of mesh size and polynomial degree. For vector finite element problems in H(curl) and H(div) (e.g., for electromagnetic or radiation diffusion problems), a specially constructed interpolation–histopolation basis is used to ensure fast convergence. Detailed performance studies are carried out to analyze the efficiency of the GPU algorithms. The kernel throughput of each of the main algorithmic components is measured, and the strong and weak parallel scalability of the methods is demonstrated. The different relative weighting and significance of the algorithmic components on GPUs and CPUs is discussed. Results on problems involving adaptively refined nonconforming meshes are shown, and the use of the preconditioners on a large-scale magnetic diffusion problem using all spaces of the finite element de Rham complex is illustrated.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Comparison of nested geometry treatments within GPU-based Monte Carlo neutron transport simulations of fission reactors

Monte Carlo (MC) neutron transport provides detailed estimates of radiological quantities within fission reactors. This involves tracking individual neutrons through a computational geometry. CPU-based MC codes use multiple polymorphic tracker types with different tracking algorithms to exploit the repeated configurations of reactors, but virtual function calls have high overhead on the GPU. The Shift MC code was modified to support GPU-based tracking with three strategies: dynamic polymorphism with virtual functions, static polymorphism, and a single tracker type with tree-based acceleration. On the Frontier supercomputer these methods achieve 77.8%, 91.2%, and 83.4%, respectively, of the tracking rate obtained using a specialized tracker optimized for rectilinear-grid-based reactors. This indicates that all three methods are suitable for typical reactor problems in which tracking does not dominate runtime. The flexibility of the single tracker method is highlighted with a hexagonal-grid microreactor problem, performed without hexagonal-grid-specific tracking routines, providing a 2.19× speedup over CPU execution.

97 MATHEMATICS AND COMPUTING↗