Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Multi-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.

83 records · Page 5
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

High-performance computing in water resources hydrodynamics

In this work, we present a vision of future water resources hydrodynamics codes that can fully utilize the strengths of modern high-performance computing. The advances to computing power, formerly driven by the improvement of central processing unit processors, now focus on parallel computing and, in particular, the use of graphics processing units (GPUs). However, this shift to a parallel framework requires refactoring the code to make efficient use of the data as well as changing even the nature of the algorithm that solves the system of equations. These concepts along with other features such as the precision for the computations, dry regions management, and input/output data are analyzed in this paper. A 2D multi-GPU flood code applied to a large-scale test case is used to corroborate our statements and ascertain the new challenges for the next-generation parallel water resources codes.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2020

Early Application Results on Pre-exascale Architecture with Analysis of Performance Challenges and Projections (Milestone PM-AD-1080 WBS 2.2)

This Exascale Computing Project (ECP) Milestone Report summarizes the status of all 30 ECP Applications Development (AD) sub-projects at the end of FY19. In August and September of 2019, a comprehensive assessment of AD projects was conducted jointly by the ECP leadership and a team of external subject matter experts. Reviews took place in person over five days-two at the National Renewable Energy Laboratory and three at Argonne National Laboratory and the University of Chicago. The review committees were tasked with evaluating each sub-project's progress in porting their code(s) to current multi-GPU architectures considered precursors to planned exascale machines. This includes characterizing which modules have been ported to multi-accelerator nodes, initial performance analyses, the status of software integration, and a current vision of successes, obstacles, and next steps. As such this report contains not only an accurate snapshot of each sub-project's current status, but also represents an unprecedentedly broad account of experiences porting large scientific applications to next-generation HPC architectures.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Accelerate M-TIP on GPUs and deploy to Summit and NERSC-9 (against simulated data) WBS 2.2.4.05 ExaFEL, Milestone ADSE13-199

We evaluated the orientation matching step in the M-TIP SPI workflow for potential offloading to accelerators. We ported the code to GPUs, benchmarked it, optimized and down-selected the best versions. The accelerated version of the orientation matching code that was developed at LANL (LANL GPU v3) is 34-55X faster than sequential, 2.4-4.9X faster than the fastest OpenMP open source version we found (FAISS OpenMP) and 1.5-4X faster than the fastest GPU open source version we found (FAISS GPU). Summit single-node GPU versions were somewhat faster than Cori GPU. Image size plays a role; mid-range image sizes take more time. The LANL CUDA multi-node, multi-GPU implementation shows mostly linear strong scaling. I/O also plays a large role; splitting data into parts improves read time and burst buffers dramatically improve read times. This work will be integrated into the M-TIP workflow as part of the next milestone ADSE13-193.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Addressing GPU memory limitations for Graph Neural Networks in High-Energy Physics applications

Introduction Reconstructing low-level particle tracks in neutrino physics can address some of the most fundamental questions about the universe. However, processing petabytes of raw data using deep learning techniques poses a challenging problem in the field of High Energy Physics (HEP). In the Exa.TrkX Project, an illustrative HEP application, preprocessed simulation data is fed into a state-of-art Graph Neural Network (GNN) model, accelerated by GPUs. However, limited GPU memory often leads to Out-of-Memory (OOM) exceptions during training, due to the large size of models and datasets. This problem is exacerbated when deploying models on High-Performance Computing (HPC) systems designed for large-scale applications. Methods We observe a high workload imbalance issue during GNN model training caused by the irregular sizes of input graph samples in HEP datasets, contributing to OOM exceptions. We aim to scale GNNs on HPC systems, by prioritizing workload balance in graph inputs while maintaining model accuracy. Our paper introduces diverse balancing strategies aimed at decreasing the maximum GPU memory footprint and avoiding the OOM exception, across various datasets. Results Our experiments showcase memory reduction of up to 32.14% compared to the baseline. We also demonstrate the proposed strategies can avoid OOM in application. Additionally, we create a distributed multi-GPU implementation using these samplers to demonstrate the scalability of these techniques on the HEP dataset. Discussion By assessing the performance of these strategies as data loading samplers across multiple datasets, we can gauge their effectiveness in both single-GPU and distributed environments. Our experiments, conducted on datasets of varying sizes and across multiple GPUs, broaden the applicability of our work to various GNN applications that handle input datasets with irregular graph sizes.

Lee, Claire Songhyun↗
DOE OSTI2022

CLAIRE—Parallelized Diffeomorphic Image Registration for Large-Scale Biomedical Imaging Applications

We study the performance of CLAIRE—a diffeomorphic multi-node, multi-GPU image-registration algorithm and software—in large-scale biomedical imaging applications with billions of voxels. At such resolutions, most existing software packages for diffeomorphic image registration are prohibitively expensive. As a result, practitioners first significantly downsample the original images and then register them using existing tools. Our main contribution is an extensive analysis of the impact of downsampling on registration performance. We study this impact by comparing full-resolution registrations obtained with CLAIRE to lower resolution registrations for synthetic and real-world imaging datasets. Our results suggest that registration at full resolution can yield a superior registration quality—but not always. For example, downsampling a synthetic image from 10243 to 2563 decreases the Dice coefficient from 92% to 79%. However, the differences are less pronounced for noisy or low contrast high resolution images. CLAIRE allows us not only to register images of clinically relevant size in a few seconds but also to register images at unprecedented resolution in reasonable time. The highest resolution considered are CLARITY images of size 2816×3016×1162. To the best of our knowledge, this is the first study on image registration quality at such resolutions.

Himthani, Naveen↗
DOE OSTI2020

Performance Evaluation of Python Based Data Analytics Frameworks in Summit: Early Experiences

The explosion in the volumes of data generated from ever-larger simulation campaigns and experiments or observations necessitates competent tools for data wrangling and analysis). While the Oak Ridge Leadership Computing Facility (OLCF) provides a variety of tools to perform data wrangling and data analysis tasks, Python based tools often lack scalability, or the ability to fully exploit the computational capability of OLCF’s Summit supercomputer. NVIDIA RAPIDS and Dask offer a promising solution to accelerate and distribute data analytics workloads from personal computers to heterogeneous supercomputing systems. We discuss early performance evaluation results of RAPIDS and Dask on Summit to understand their capabilities, scalability, and limitations. Our evaluation includes a subset of RAPIDS libraries, i.e., cuDF, cuML, and cuGraph, and Chainer’s CuPy, and their multi-GPU variants when available.We also draw on the observed trends from the performance evaluation results to discuss best practices for maximizing performance.

Hernandez Arreguin, Benjamin↗
DOE OSTI2020

C-SAW: a framework for graph sampling and random walk on GPUs

Many applications require to learn, mine, analyze and visualize large-scale graphs. These graphs are often too large to be addressed efficiently using conventional graph processing technologies. Fortunately, recent research efforts find out graph sampling and random walk, which significantly reduce the size of original graphs, can benefit the tasks of learning, mining, analyzing and visualizing large graphs by capturing the desirable graph properties. This paper introduces C-SAW, the first framework that accelerates Sampling and Random Walk framework on GPUs. Particularly, C-SAW makes three contributions: First, our framework provides a generic API which allows users to implement a wide range of sampling and random walk algorithms with ease. Second, offloading this framework on GPU, we introduce warp-centric parallel selection, and two novel optimizations for collision migration. Third, towards supporting graphs that exceed the GPU memory capacity, we introduce efficient data transfer optimizations for out-of-memory and multi-GPU sampling, such as workload-aware scheduling and batched multi-instance sampling. Taken together, our framework constantly outperforms the state of the art projects in addition to the capability of supporting a wide range of sampling and random walk algorithms.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

OPER: Optimality-Guided Embedding Table Parallelization for Large-scale Recommendation Model

With the sharp increasing volume of user data, Deep Learning Recommendation Model (DLRM) becomes an indispensable infrastructure in large technology companies. However, large-scale DLRM on the multi-GPU platform is still inefficient due to unbalanced workload partitioning and intensive inter-GPU communication. To this end, we propose OPER, an OPtimality guided Embedding table placement for large-scale Recommendation model training and inference. OPER explores the potential of mitigating remote memory access latency in DLRM through fine-grained embedding table placement. Specifically, OPER proposes a theoretical modeling that builds up the relationship between EMT placement and the embedding communication latency in both training and inference. OPER proves the NP hardness of finding the optimal embedding table placement and proposes a heuristic algorithm that yields near optimal placement. OPER implements a SHMEM-based embedding table training system and a unified embedding index mapping to support fine-grained embedding table sharding and placement. Comprehensive experiments reveal that OPER achieves on average 3.4× and 5.1× speedup on training and inference respectively over state-of-the-art DLRM frameworks.

Wang, Zheng↗
DOE OSTI2021

Parallel physics-informed neural networks via domain decomposition

Here we develop a distributed framework for the physics-informed neural networks (PINNs) based on two recent extensions, namely conservative PINNs (cPINNs) and extended PINNs (XPINNs), which employ domain decomposition in space and in time-space, respectively. This domain decomposition endows cPINNs and XPINNs with several advantages over the vanilla PINNs, such as parallelization capacity, large representation capacity, efficient hyperparameter tuning, and is particularly effective for multi-scale and multi-physics problems. Here, we present a parallel algorithm for cPINNs and XPINNs constructed with a hybrid programming model described by MPI + X, where X ∈ {CPUs, GPUs}. The main advantage of cPINN and XPINN over the more classical data and model parallel approaches is the flexibility of optimizing all hyperparameters of each neural network separately in each subdomain. We compare the performance of distributed cPINNs and XPINNs for various forward problems, using both weak and strong scalings. Our results indicate that for space domain decomposition, cPINNs are more efficient in terms of communication cost but XPINNs provide greater flexibility as they can also handle time-domain decomposition for any differential equations, and can deal with any arbitrarily shaped complex subdomains. To this end, we also present an application of the parallel XPINN method for solving an inverse diffusion problem with variable conductivity on the United States map, using ten regions as subdomains.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2022

A Block-Based Triangle Counting Algorithm on Heterogeneous Environments

Triangle counting is a fundamental building block in graph algorithms. In this article, we propose a block-based triangle counting algorithm to reduce data movement during both sequential and parallel execution. Our block-based formulation makes the algorithm naturally suitable for heterogeneous architectures. The problem of partitioning the adjacency matrix of a graph is well-studied. Our task decomposition goes one step further: it partitions the set of triangles in the graph. By streaming these small tasks to compute resources, we can solve problems that do not fit on a device. We demonstrate the effectiveness of our approach by providing an implementation on a compute node with multiple sockets, cores and GPUs. The current state-of-the-art in triangle enumeration processes the Friendster graph in 2.1 seconds, not including data copy time between CPU and GPU. Using that metric, our approach is 20 percent faster. When copy times are included, our algorithm takes 3.2 seconds. This is 5.6 times faster than the fastest published CPU-only time.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

A Block-Based Triangle Counting Algorithm on Heterogeneous Environments

Triangle counting is a fundamental building block in graph algorithms. In this paper, we propose a block-based triangle counting algorithm to reduce data movement during both sequential and parallel execution. Our block-based formulation makes the algorithm naturally suitable for heterogeneous architectures. The problem of partitioning the adjacency matrix of a graph is well-studied. Our task decomposition goes one step further: it partitions the set of triangles in the graph. By streaming these small tasks to compute resources, we can solve problems that do not fit on a device. We demonstrate the effectiveness of our approach by providing an implementation on a compute node with multiple sockets, cores and GPUs. The current state-of-the-art in triangle enumeration processes the Friendster graph in 2.1 seconds, not including data copy time between CPU and GPU. Using that metric, our approach is 20 percent faster. When copy times are included, our algorithm takes 3.2 seconds. This is 5.6 times faster than the fastest published CPU-only time.

97 MATHEMATICS AND COMPUTING↗