Search NASA⌕ Search

SEARCH · Search NASA

Results for “GPU accelerators”

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 595 records · Page 33

Enabling real-time adaptation of machine learning models at x-ray Free Electron Laser facilities with high-speed training optimized computational hardware

The emergence of novel computational hardware is enabling a new paradigm for rapid machine learning model training. For the Department of Energy’s major research facilities, this developing technology will enable a highly adaptive approach to experimental sciences. In this manuscript we present the per-epoch and end-to-end training times for an example of a streaming diagnostic that is planned for the upcoming high-repetition rate x-ray Free Electron Laser, the Linac Coherent Light Source-II. We explore the parameter space of batch size and data parallel training across multiple Graphics Processing Units and Reconfigurable Dataflow Units. We show the landscape of training times with a goal of full model retraining in under 15 min. Although a full from scratch retraining of a model may not be required in all cases, we nevertheless present an example of the application of emerging computational hardware for adapting machine learning models to changing environments in real-time, during streaming data acquisition, at the rates expected for the data fire hoses of accelerator-based user facilities.

97 MATHEMATICS AND COMPUTING↗

Extreme-bandwidth scalable performance-per-watt GPU architecture

A technique for accessing memory in an accelerated processing device coupled to stacked memory dies is provided herein. The technique includes receiving a memory access request from an execution unit and identifying whether the memory access request corresponds to memory cells of the stacked dies that are considered local to the execution unit or non-local. For local accesses, the access is made “directly”, that is, without using a bus. A control die coordinates operations for such local accesses, activating particular through-silicon-vias associated with the memory cells that include the data for the access. Non-local accesses are made via a distributed cache fabric and an interconnect bus in the control die. Various other features and details are provided below.

97 MATHEMATICS AND COMPUTING↗

Efficient Parallelization of Irregular Applications on GPU Architectures

With the enlarging computation capacity of general Graphics Processing Units (GPUs), leveraging GPUs to accelerate parallel applications has become a critical topic in academia and industry. However, a wide range of irregular applications with the computation-/memory-intensive nature cannot easily achieve high GPU utilization. The challenges mainly involve the following aspects: first, data dependence leads to coarse-grained kernel and inefficient parallelism; second, heavy GPU memory usage may cause frequent memory evictions and extra overhead of I/O; third, specific computation patterns produce memory redundancies; last, workload balance and data reusability conjunctly benefit the overall performance, but there may exist a dynamic trade-off between them. Targeting these challenges, this dissertation proposes multiple optimizations to accelerate two real-world applications: many-body correlation functions to simulate nuclear physics in a large-scale scientific system; the other is the eALS-based matrix factorization recommendation system. To accelerate the calculations of many-body correlation functions, this dissertation presents three frameworks in GPU memory management and multi-GPU scheduling. Firstly, an optimized systematic GPU memory management framework, MemHC, utilizes a series of new memory reduction designs in GPU memory allocation, CPU/GPU communications, and GPU memory oversubscription. Secondly, an enhanced multi-GPU scheduling framework, MICCO, particularly by taking both data dimension (e.g., data reuse and data eviction) and computation dimension into account. MICCO designs a heuristic scheduling algorithm and a machine learning-based regression model to generate the optimal settings of a proposed new concept to manage the trade-off. Thirdly, a locality-aware multi-GPU scheduling framework. This scheduler leverages pipeline batch generation with a looking-ahead strategy by building local dependency graphs for memory transfer reduction and better data reuse, achieving up to 79.92% memory cost reduction and 1.67x speedup. To parallelize the eALS-based recommendation system, this dissertation proposes an efficient CPU/GPU heterogeneous recommendation system, HEALS. HEALS employs newly designed architecture-adaptive data formats to achieve load balance and good data locality on CPU and GPU. To mitigate the data dependence, HEALS presents a CPU/GPU collaboration model for both task parallelism and data parallelism with multiple kernel computation optimizations. In summary, this dissertation efficiently accelerates two typical irregular applications on GPUs by building four frameworks, including CPU/GPU collaboration, GPU memory management, and multi-GPU scheduling.

Wang, Qihan↗

DRAGON: breaking GPU memory capacity limits with direct NVM access

Heterogeneous computing with accelerators is growing in importance in high performance computing (HPC). Recently, application datasets have expanded beyond the memory capacity of these accelerators, and often beyond the capacity of their hosts. Meanwhile, nonvolatile memory (NVM) storage has emerged as a pervasive component in HPC systems because NVM provides massive amounts of memory capacity at affordable cost. Currently, for accelerator applications to use NVM, they must manually orchestrate data movement across multiple memories and this approach only performs well for applications with simple access behaviors. To address this issue, we developed DRAGON, a solution that enables all classes of GP-GPU applications to transparently compute on terabyte datasets residing in NVM. DRAGON leverages the page-faulting mechanism on the recent NVIDIA GPUs by extending capabilities of CUDA Unified Memory (UM). Our experimental results show that DRAGON transparently expands memory capacity and obtain additional speedups via automated I/O and data transfer overlapping.

Markthub, Pak↗

Kernel fusion in atomistic spin dynamics simulations on Nvidia GPUs using tensor core

In atomistic spin dynamics simulations, the time cost of constructing the space- and time-displaced pair correlation function in real space increases quadratically as the number of spins N, leading to significant computational effort. The GEMM subroutine can be adopted to accelerate the calculation of the dynamical spin-spin correlation function, but the computational cost of simulating large spin systems (>40000 spins) on CPUs remains expensive. In this work, we perform the simulation on the graphics processing unit (GPU), a hardware solution widely used as an accelerator for scientific computing and deep learning. Here we show that GPUs can accelerate the simulation up to 25-fold compared to multi-core CPUs when using the GEMM subroutine on both. To hide memory latency, we fuse the element-wise operation into the GEMM kernel using CUTLASS that can improve the performance by 26% ~ 33% compared to implementation based on cuBLAS. Furthermore, we perform the on-the-fly calculation in the epilogue of the GEMM subroutine to avoid saving intermediate results on global memory, which makes the large-scale atomistic spin dynamics simulation feasible and affordable.

97 MATHEMATICS AND COMPUTING↗

Towards Efficient Alternating Current Optimal Power Flow Analysis on Graphical Processing Units

We present a solution of sparse ACOPF analysis on GPU. In particular, we discuss the performance bottlenecks and detail our efforts to accelerate the linear solver, a core component of ACOPF that dominates the computational time. ACOPF solutions of two large-scale systems, synthetic Northeast (25,000 buses) and Eastern (70,000 buses) \cite{birchfield2017tamu-cases} on GPU show promising speed-up compared to CPU based solution using a state-of-the-art solver. To our knowledge, this is the first result demonstrating acceleration of sparse ACOPF on GPUs.

Power grid analysis, GPU↗

General field evaluation in high-order meshes on GPUs

Robust and scalable function evaluation at any arbitrary point in the finite/spectral element mesh is required for querying the partial differential equation solution at points of interest, comparison of solution between different meshes, and Lagrangian particle tracking. This is a challenging problem, particularly for high-order unstructured meshes partitioned in parallel with MPI, as it requires identifying the element that overlaps a given point and computing the corresponding reference space coordinates. Here, we present a robust and efficient technique for general field evaluation in large-scale high-order meshes with quadrilaterals and hexahedra. In the proposed method, a combination of globally partitioned and processor-local maps are used to first determine a list of candidate MPI ranks, and then locally candidate elements that could contain a given point. Next, element-wise bounding boxes further reduce the list of candidate elements. Finally, Newton’s method with trust region is used to determine the overlapping element and corresponding reference space coordinates. Since GPU-based architectures have become popular for accelerating computational analyses using meshes with tensor-product elements, specialized kernels have been developed to utilize the proposed methodology on GPUs. The method is also extended to enable general field evaluation on surface meshes. The paper concludes by demonstrating the use of the proposed method in various applications ranging from mesh-to-mesh transfer during r-adaptivity to Lagrangian particle tracking.

97 MATHEMATICS AND COMPUTING↗

New Horizons for High-Performance Computing

Here we provide an overview of the past, present, and a diverse collection of future computer architecture alternatives for HPC. The end of Moore’s Law influenced the current HPC architecture focus on accelerated compute nodes composed of CPU and GPU computing components integrated into massively parallel processor architecture systems. There are many alternatives for future HPC directions, with different technologies, computing ecosystems, opportunities for lead user application-driven customization, and the role of open innovation business models. This paper provides an overview of these different new horizons for HPC, an organizing principle to focus future computing research, different public-private partnership models, and the critical role of workforce development.

97 MATHEMATICS AND COMPUTING↗

VTK-m: Visualization for the Exascale Era and Beyond

A recent trend in modern high-performance computing is the increasing use of hybrid architectures, where the vast majority of performance comes from accelerators. Modern accelerators are based on Graphics Processing Units (GPU) that contain many low power cores that in their aggregate provides an extremely high computation rate. Current and future CPU processors are requiring more explicit parallelism as each successive version of the hardware packs in more cores, and technologies like hyperthreading and vector operations require even more parallel processing to leverage each core’s full potential. As an example, the Frontier supercomputer installed at Oak Ridge National Laboratories recently hit a record breaking 1.1 exaflops1 on the LINPACK HPC benchmark [Shoemaker 2022]. The system contains 37632 AMD MI250x GPUs which requires more than half a billion threads to keep the system fully utilized [Khizeran 2022].VTK-m is a toolkit of scientific visualization algorithms for these emerging processor architectures. VTK-m supports the fine-grained concurrency for data analysis and visualization algorithms required to drive extreme scale computing by providing abstract models for data and execution that can be applied to a variety of algorithms across many different processor architectures.

Bolstad, Mark↗

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↗

Develop a Fast Analysis Solver for Welding Sequence Optimization

During the shipbuilding manufacturing process, materials are exposed to significant stresses, as induced both thermally and mechanically, that alter the intended design and significantly affect the production schedule, labor hours (fitting, welding, rework, etc.), and material structural performance. The type and magnitude of deformation of a given structure depends on many factors such as the material, thickness and quality of components, the process heat input, preheat and inter-pass temperatures, type and size of welds, welding sequence and direction, location, sequence, and degree of fixturing. Numerical simulations using finite element analysis (FEA) have long been used to analyze welding-induced structural distortion. For large assemblies, transient thermal elastic-plastic analysis (TEPA) can take days or weeks to run, and optimization of welding sequence is not feasible. Simplified analysis methods were developed to reduce computational time. However, it is challenging to use these techniques to fully optimize welding sequencing because of their applied simplifications in modeling weld details. A fast analysis solver that could be used by the shipbuilding industry is being developed for optimizing welding sequences by taking full advantage of modern GPU-based HPC hardware and incorporating patented acceleration schemes. The accelerated processing factors are up to 2200 times greater for large, multi-pass welded structures.

Yang, Yu-Ping↗

Toward large-scale simulation of residual stress and distortion in wire and arc additive manufacturing

In this study, we aim to advance the structural analysis of wire and arc additive manufacturing (WAAM) by considering the thermomechanical features inherent in direct energy deposition. Simulation approaches including the iterative substructure method (ISM), dynamic mesh refining method (DMRM), and graphics processing unit (GPU) based explicit finite element method (FEM) were developed for accelerating additive manufacturing stress analysis that is very time consuming by conventional numerical methods. ISM and DMRM take advantage of the strong nonlinearity phenomenon near the moving heat source by reducing the global iterations and refining the local mesh, respectively. In addition to the use of GPUs, the explicit FEM is accelerated by a time scaling technique based on the inherent strain concept. The residual stress and distortion of two large builds were analyzed, showing very consistent numerical results and good agreement with experiments. Compared with the commercial software Abaqus, the novel approaches reduced the computational cost substantially without compromising accuracy. The GPU code showed the highest computational efficiency (30~70×), while DMRM and ISM had acceleration factors of 9× and 3×, respectively. Finally, such high-fidelity modeling approaches will be very useful for building up a digital twin of WAAM to reduce development time and cost.

36 MATERIALS SCIENCE↗

GPU coprocessors as a service for deep learning inference in high energy physics

In the next decade, the demands for computing in large scientific experiments are expected to grow tremendously. During the same time period, CPU performance increases will be limited. At the CERN Large Hadron Collider (LHC), these two issues will confront one another as the collider is upgraded for high luminosity running. Alternative processors such as graphics processing units (GPUs) can resolve this confrontation provided that algorithms can be sufficiently accelerated. In many cases, algorithmic speedups are found to be largest through the adoption of deep learning algorithms. We present a comprehensive exploration of the use of GPU-based hardware acceleration for deep learning inference within the data reconstruction workflow of high energy physics. We present several realistic examples and discuss a strategy for the seamless integration of coprocessors so that the LHC can maintain, if not exceed, its current performance throughout its running.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Enabling Scientific Applications with Performance-Portability and High-Productivity for Multi-GPU Programming with JACC.Multi

This work bridges the gap between multi-GPU computing and high-productivity, performance-portable programming solutions. Our goal is to enhance scientific applications with a productive and portable solution—program once, deploy everywhere—for multi-GPU programming with no cost to programmability. To accomplish this, we implemented JACC.Multi, which is part of the Julia for ACCelerators (JACC) performance-portable framework. JACC. Multi is the only high-level, portable metaprogramming solution that targets multi-GPU environments and is integrated in a readily accessible programming language (e.g., Julia language). With transparent GPU-to-GPU communication, JACC. Multi is optimized for scientific application workloads and is portable for NVIDIA and AMD accelerators. For the evaluation, we use two modern multi-GPU systems: Hudson, which features two NVIDIA H100 Hopper GPUs per node, and Frontier, which features four AMD MI250X GPUs per node, each with two Graphics Compute Dies (GCDs) for a total of eight GCDs per node. Additionally, as part of the evaluation, we use JACC (one GPU), MPI+JACC, and JACC. Multi codes that implement well-known and widely used scientific algorithms/kernels such as the conjugate gradient algorithm and an explicit forward Euler solver that requires GPU-to-GPU communication. Overall, JACC. Multi codes achieve better performance than MPI+JACC codes and significant speedups over JACC (one GPU), with up to 1.9× on Hudson and 6× on Frontier.

Valero Lara, Pedro [ORNL] (ORCID:0000000214794310)↗

Frog: Asynchronous Graph Processing on GPU with Hybrid Coloring Model

GPUs have been increasingly used to accelerate graph processing for complicated computational problems regarding graph theory. Many parallel graph algorithms adopt the asynchronous computing model to accelerate the iterative convergence. Unfortunately, the consistent asynchronous computing requires locking or atomic operations, leading to significant penalties/overheads when implemented on GPUs. As such, coloring algorithm is adopted to separate the vertices with potential updating conflicts, guaranteeing the consistency/correctness of the parallel processing. Common coloring algorithms, however, may suffer from low parallelism because of a large number of colors generally required for processing a large-scale graph with billions of vertices. We propose a light-weight asynchronous processing framework called Frog with a preprocessing/hybrid coloring model. The fundamental idea is based on Pareto principle (or 80-20 rule) about coloring algorithms as we observed through masses of realworld graph coloring cases. We find that a majority of vertices (about 80%) are colored with only a few colors, such that they can be read and updated in a very high degree of parallelism without violating the sequential consistency. Accordingly, our solution separates the processing of the vertices based on the distribution of colors. In this work, we mainly answer three questions: (1) how to partition the vertices in a sparse graph with maximized parallelism, (2) how to process large-scale graphs that cannot fit into GPU memory, and (3) how to reduce the overhead of data transfers on PCIe while processing each partition. Here, we conduct experiments on real-world data (Amazon, DBLP, YouTube, RoadNet-CA, WikiTalk and Twitter) to evaluate our approach and make comparisons with well-known non-preprocessed (such as Totem, Medusa, MapGraph and Gunrock) and preprocessed (Cusha) approaches, by testing four classical algorithms (BFS, PageRank, SSSP and CC). On all the tested applications and datasets, Frog is able to significantly outperform existing GPU-based graph processing systems except Gunrock and MapGraph. MapGraph gets better performance than Frog when running BFS on RoadNet-CA. The comparison between Gunrock and Frog is inconclusive. Frog can outperform Gunrock more than 1.04X when running PageRank and SSSP, while the advantage of Frog is not obvious when running BFS and CC on some datasets especially for RoadNet-CA.

97 MATHEMATICS AND COMPUTING↗

UCLA Final Technical Report for the "Community Petascale Project for Accelerator Science and Simulation”.

The UCLA Plasma Simulation Group is a major partner of the “Community Petascale Project for Accelerator Science and Simulation”. This is the final technical report. We include an overall summary, a list of publications, progress for the most recent year, and individual progress reports for each year. We have made tremendous progress during the three years. SciDAC funds have contributed to the development of a large number of skeleton codes that illustrate how to write PIC codes with a hierarchy of parallelism. These codes cover 2D and 3D as well as electrostatic solvers (which are used in beam dynamics codes and quasi-static codes) and electromagnetic solvers (which are used in plasma based accelerator codes). We also used these ideas to develop a GPU enabled version of OSIRIS. SciDAC funds were also contributed to the development of strategies to eliminate the Numerical Cerenkov Instability (NCI) which is an issue when carrying laser wakefield accelerator (LWFA) simulations in a boosted frame and when quantifying the emittance and energy spread of self-injected electron beams. This work included the development of a new code called UPIC-EMMA which is an FFT based electromagnetic PIC code and to new hybrid algorithms in OSIRIS. A new hybrid (PIC in r-z and gridless in φ) algorithm was implemented into OSIRIS. In this algorithm the fields and current are expanded into azimuthal harmonics and the complex amplitude for each harmonic is calculated separately. The contributions from each harmonic are summed and then used to push the particles. This algorithm permits modeling plasma based acceleration with some 3D effects but with the computational load of an 2D r-z PIC code. We developed a rigorously charge conserving current deposit for this algorithm. Very recently, we made progress in combining the speed up from the quasi-3D algorithm with that from the Lorentz boosted frame. SciDAC funds also contributed to the improvement and speed up of the quasi-static PIC code QuickPIC. We have also used our suite of PIC codes to make scientific discovery. Highlights include supporting FACET experiments which achieved the milestones of showing high beam loading and energy transfer efficiency from a drive electron beam to a witness electron beam and the discovery of a self-loading regime a for high gradient acceleration of a positron beam. Both of these experimental milestones were published in Nature together with supporting QuickPIC simulation results. Simulation results from QuickPIC were used on the cover of Nature in one case. We are also making progress on using highly resolved QuickPIC simulations to show that ion motion may not lead to catastrophic emittance growth for tightly focused electron bunches loaded into nonlinear wakefields. This could mean that fully self-consistent beam loading scenarios are possible. This work remains in progress. OSIRIS simulations were used to discover how 200 MeV electron rings are formed in LWFA experiments, on how to generate electrons that have a series of bunches on nanometer scale, and how to transport electron beams from (into) plasma sections into (from) conventional beam optic sections.

43 PARTICLE ACCELERATORS↗

A graphics processing unit accelerated sparse direct solver and preconditioner with block low rank compression

We present the GPU implementation efforts and challenges of the sparse solver package STRUMPACK. The code is made publicly available on github with a permissive BSD license. STRUMPACK implements an approximate multifrontal solver, a sparse LU factorization which makes use of compression methods to accelerate time to solution and reduce memory usage. Multiple compression schemes based on rank-structured and hierarchical matrix approximations are supported, including hierarchically semi-separable, hierarchically off-diagonal butterfly, and block low rank. Here, in this paper, we present the GPU implementation of the block low rank (BLR) compression method within a multifrontal solver. Our GPU implementation relies on highly optimized vendor libraries such as cuBLAS and cuSOLVER for NVIDIA GPUs, rocBLAS and rocSOLVER for AMD GPUs and the Intel oneAPI Math Kernel Library (oneMKL) for Intel GPUs. Additionally, we rely on external open source libraries such as SLATE (Software for Linear Algebra Targeting Exascale), MAGMA (Matrix Algebra on GPU and Multi-core Architectures), and KBLAS (KAUST BLAS). SLATE is used as a GPU-capable ScaLAPACK replacement. From MAGMA we use variable sized batched dense linear algebra operations such as GEMM, TRSM and LU with partial pivoting. KBLAS provides efficient (batched) low rank matrix compression for NVIDIA GPUs using an adaptive randomized sampling scheme. The resulting sparse solver and preconditioner runs on NVIDIA, AMD and Intel GPUs. Interfaces are available from PETSc, Trilinos and MFEM, or the solver can be used directly in user code. We report results for a range of benchmark applications, using the Perlmutter system from NERSC, Frontier from ORNL, and Aurora from ALCF. For a high frequency wave equation on a regular mesh, using 32 Perlmutter compute nodes, the factorization phase of the exact GPU solver is about 6.5× faster compared to the CPU-only solver. The BLR-enabled GPU solver is about 13.8× faster than the CPU exact solver. For a collection of SuiteSparse matrices, the STRUMPACK exact factorization on a single GPU is on average 1.9× faster than NVIDIA’s cuDSS solver.

97 MATHEMATICS AND COMPUTING↗