Search NASA⌕ Search

SEARCH · Search NASA

Results for “GRAPHICS PROCESSING UNITS”

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 91 records · Page 5

Resident Block-Structured Adaptive Mesh Refinement on Thousands of Graphics Processing Units, In: 2015 44th International Conference on Parallel Processing

Block-structured adaptive mesh refinement (AMR) is a technique that can be used when solving partial differential equations to reduce the number of cells necessary to achieve the required accuracy in areas of interest. These areas (shock fronts, material interfaces, etc.) are recursively covered with finer mesh patches that are grouped into a hierarchy of refinement levels. Despite the potential for large savings in computational requirements and memory usage without a corresponding reduction in accuracy, AMR adds overhead in managing the mesh hierarchy, adding complex communication and data movement requirements to a simulation. In this paper, we describe the design and implementation of a resident GPU-based AMR library, including: the classes used to manage data on a mesh patch, the routines used for transferring data between GPUs on different nodes, and the data-parallel operators developed to coarsen and refine mesh data. We validate the performance and accuracy of our implementation using three test problems and two architectures: an 8 node cluster, and 4,196 nodes of Oak Ridge National Laboratory's Titan supercomputer. Our GPU-based AMR hydrodynamics code performs up to 4.87× faster than the CPU-based implementation, and is scalable on 4,196 K20x GPUs using a combination of MPI and CUDA.

Computer Science↗

Tensor Algebra Library for NVidia Graphics Processing Units

This is a general purpose math library implementing basic tensor algebra operations on NVidia GPU accelerators. This software is a tensor algebra library that can perform basic tensor algebra operations, including tensor contractions, tensor products, tensor additions, etc., on NVidia GPU accelerators, asynchronously with respect to the CPU host. It supports a simultaneous use of multiple NVidia GPUs. Each asynchronous API function returns a handle which can later be used for querying the completion of the corresponding tensor algebra operation on a specific GPU. The tensors participating in a particular tensor operation are assumed to be stored in local RAM of a node or GPU RAM. The main research area where this library can be utilized is the quantum many-body theory (e.g., in electronic structure theory).

Liakh, Dmitry↗

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↗

Parallel Block Structured Adaptive Mesh Refinement on Graphics Processing Units

Block-structured adaptive mesh refinement is a technique that can be used when solving partial differential equations to reduce the number of zones necessary to achieve the required accuracy in areas of interest. These areas (shock fronts, material interfaces, etc.) are recursively covered with finer mesh patches that are grouped into a hierarchy of refinement levels. Despite the potential for large savings in computational requirements and memory usage without a corresponding reduction in accuracy, AMR adds overhead in managing the mesh hierarchy, adding complex communication and data movement requirements to a simulation. In this paper, we describe the design and implementation of a native GPU-based AMR library, including: the classes used to manage data on a mesh patch, the routines used for transferring data between GPUs on different nodes, and the data-parallel operators developed to coarsen and refine mesh data. We validate the performance and accuracy of our implementation using three test problems and two architectures: an eight-node cluster, and over four thousand nodes of Oak Ridge National Laboratory’s Titan supercomputer. Our GPU-based AMR hydrodynamics code performs up to 4.87× faster than the CPU-based implementation, and has been scaled to over four thousand GPUs using a combination of MPI and CUDA.

97 MATHEMATICS AND COMPUTING↗

Ground-referenced single-ended signaling connected graphics processing unit multi-chip module

A system of interconnected chips comprising a multi-chip module (MCM) includes a processor chip, a system functions chip, and an MCM package configured to include the processor chip, the system functions chip, and an interconnect circuit. The processor chip is configured to include a first ground-referenced single-ended signaling interface circuit. A first set of electrical traces manufactured within the MCM package and configured to couple the first single-ended signaling interface circuit to the interconnect circuit. The system functions chip is configured to include a second single-ended signaling interface circuit and a host interface. A second set of electrical traces manufactured within the MCM package and configured to couple the host interface to at least one external pin of the MCM package. In one embodiment, each single-ended signaling interface advantageously implements ground-referenced single-ended signaling.

Dally, William J.↗

Heterogeneous graphics processing unit for scheduling thread groups for execution on variable width SIMD units

A compute unit configured to execute multiple threads in parallel is presented. The compute unit includes one or more single instruction multiple data (SIMD) units and a fetch and decode logic. The SIMD units have differing numbers of arithmetic logic units (ALUs), such that each SIMD unit can execute a different number of threads. The fetch and decode logic is in communication with each of the SIMD units, and is configured to assign the threads to the SIMD units for execution based on such differing numbers of ALUs.

97 MATHEMATICS AND COMPUTING↗

Hardware accelerated dynamic work creation on a graphics processing unit

A processor core is configured to execute a parent task that is described by a data structure stored in a memory. A coprocessor is configured to dispatch a child task to the at least one processor core in response to the coprocessor receiving a request from the parent task concurrently with the parent task executing on the at least one processor core. In some cases, the parent task registers the child task in a task pool and the child task is a future task that is configured to monitor a completion object and enqueue another task associated with the future task in response to detecting the completion object. The future task is configured to self-enqueue by adding a continuation future task to a continuation queue for subsequent execution in response to the future task failing to detect the completion object.

Gutierrez, Anthony↗

Hardware accelerated dynamic work creation on a graphics processing unit

A processor core is configured to execute a parent task that is described by a data structure stored in a memory. A coprocessor is configured to dispatch a child task to the at least one processor core in response to the coprocessor receiving a request from the parent task concurrently with the parent task executing on the at least one processor core. In some cases, the parent task registers the child task in a task pool and the child task is a future task that is configured to monitor a completion object and enqueue another task associated with the future task in response to detecting the completion object. The future task is configured to self-enqueue by adding a continuation future task to a continuation queue for subsequent execution in response to the future task failing to detect the completion object.

Gutierrez, Anthony T.↗

Lattice QCD and Graphical Processing Units

New heavy vector bosons, also known as ``heavy photons'', ``dark photons'', or ``hidden sector photons'', are expected on very general theoretical grounds. Recent astrophysical evidence motivates the direct search for a heavy photon $A'$ in the mass range $m_{A'} \sim 20$ to 1000 MeV/c$^{2}$, with coupling to ordinary photons {\it via} a kinetic mixing term $\propto\epsilon$, expected around $10^{-5}$ to $10^{-2}$. Such a new dark force gauge boson below the GeV range would naturally mediate TeV range dark matter annihilation, and interaction with ordinary matter. Existing constraints from collider searches and beam dump experiments leave much of this preferred $(m_{A'},\epsilon)$ phase space unexplored. The detection of electroproduced $A'$ on a heavy target from their decay into $e^{+}e^{-}$ pairs comes with a copious QED trident background. Using the high luminosity and precision beam available from Jlab, and combining a silicon microstrip vertex tracker with a lead-tungstate calorimeter for triggering, the HPS experiment will hunt for the $A'$ in a large $\epsilon$ range, with unique sensitivity in the lower coupling range using vertexing to suppress the trident background. In this talk, the experimental setup and goals of the HPS experiment will be presented.

Joo, Balint↗

DOE SBIR Phase-1 Report on Hybrid CPU-GPU Parallel Development of the Eulerian-Lagrangian Barracuda Multiphase Program

This report gives the result from the Phase-1 work on demonstrating greater than 10x speedup of the Barracuda computer program using parallel methods and GPU processors (General-Purpose Graphics Processing Unit or Graphics Processing Unit). Phase-1 demonstrated a 12x speedup on a typical Barracuda function using the GPU processor. The problem test case used about 5 million particles and 250,000 Eulerian grid cells. The relative speedup, compared to a single CPU, increases with increased number of particles giving greater than 12x speedup. Phase-1 work provided a path for reformatting data structure modifications to give good parallel performance while keeping a friendly environment for new physics development and code maintenance. The implementation of data structure changes will be in Phase-2. Phase-1 laid the ground work for the complete parallelization of Barracuda in Phase-2, with the caveat that implemented computer practices for parallel programming done in Phase-1 gives immediate speedup in the current Barracuda serial running code. The Phase-1 tasks were completed successfully laying the frame work for Phase-2. The detailed results of Phase-1 are within this document. In general, the speedup of one function would be expected to be higher than the speedup of the entire code because of I/O functions and communication between the algorithms. However, because one of the most difficult Barracuda algorithms was parallelized in Phase-1 and because advanced parallelization methods and proposed parallelization optimization techniques identified in Phase-1 will be used in Phase-2, an overall Barracuda code speedup (relative to a single CPU) is expected to be greater than 10x. This means that a job which takes 30 days to complete will be done in 3 days. Tasks completed in Phase-1 are: Task 1: Profile the entire Barracuda code and select which subroutines are to be parallelized (See Section Choosing a Function to Accelerate) Task 2: Select a GPU consultant company and jointly parallelize subroutines (CPFD chose the small business EMPhotonics for the Phase-1 the technical partner. See Section Technical Objective and Approach) Task 3: Integrate parallel subroutines into Barracuda (See Section Results from Phase-1 and its subsections) Task 4: Testing, refinement, and optimization of parallel methodology (See Section Results from Phase-1 and Section Result Comparison Program) Task 5: Integrate Phase-1 parallel subroutines into Barracuda and release (See Section Results from Phase-1 and its subsections) Task 6: Roadmap of Phase-2 (See Section Plan for Phase-2) With the completion of Phase 1 we have the base understanding to completely parallelize Barracuda. An overview of the work to move Barracuda to a parallelized code is given in Plan for Phase-2.

02 PETROLEUM↗

Modeling of a chain of three plasma accelerator stages with the WarpX electromagnetic PIC code on GPUs

The fully electromagnetic particle-in-cell code WarpX is being developed by a team of the U.S. DOE Exascale Computing Project (with additional non-U.S. collaborators on part of the code) to enable the modeling of chains of tens to hundreds of plasma accelerator stages on exascale supercomputers, for future collider designs. The code is combining the latest algorithmic advances (e.g., Lorentz boosted frame and pseudo-spectral Maxwell solvers) with mesh refinement and runs on the latest computer processing unit and graphical processing unit (GPU) architectures. In this paper, we summarize the strategy that was adopted to port WarpX to GPUs, report on the weak parallel scaling of the pseudo-spectral electromagnetic solver, and then present solutions for decreasing the time spent in data exchanges from guard regions between subdomains. In Sec. IV, we demonstrate the simulations of a chain of three consecutive multi-GeV laser-driven plasma accelerator stages.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Subspace recursive Fermi-operator expansion strategies for large-scale DFT eigenvalue problems on HPC architectures

Quantum mechanical calculations for material modeling using Kohn–Sham density functional theory (DFT) involve the solution of a nonlinear eigenvalue problem for N smallest eigenvector-eigenvalue pairs, with N proportional to the number of electrons in the material system. Here, these calculations are computationally demanding and have asymptotic cubic scaling complexity with the number of electrons. Large-scale matrix eigenvalue problems arising from the discretization of the Kohn–Sham DFT equations employing a systematically convergent basis traditionally rely on iterative orthogonal projection methods, which are shown to be computationally efficient and scalable on massively parallel computing architectures. However, as the size of the material system increases, these methods are known to incur dominant computational costs through the Rayleigh–Ritz projection step of the discretized Kohn–Sham Hamiltonian matrix and the subsequent subspace diagonalization of the projected matrix. This work explores the potential of polynomial expansion approaches based on recursive Fermi-operator expansion as an alternative to the subspace diagonalization of the projected Hamiltonian matrix to reduce the computational cost. Subsequently, we perform a detailed comparison of various recursive polynomial expansion approaches to the traditional approach of explicit diagonalization on both multi-node central processing unit and graphics processing unit architectures and assess their relative performance in terms of accuracy, computational efficiency, scaling behavior, and energy efficiency.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Many-body perturbation theory with hybrid density functional theory starting points accelerated by adaptively compressed exchange

We report on the use of the adaptively compressed exchange (ACE) operator to accelerate many-body perturbation theory (MBPT) calculations, including G 0 W 0 and the Bethe–Salpeter equation (BSE), for hybrid density functional theory starting points. We show that by approximating the exact exchange operator with the low-rank ACE operator, substantial computational savings can be achieved with systematically controllable errors in the quasiparticle energies computed with full-frequency G 0 W 0 and the optical absorption spectra and vertical excitation energies computed by solving the BSE within density matrix perturbation theory. Our implementation makes use of the ACE-accelerated electronic Hamiltonian to carry out both G 0 W 0 and BSE without explicitly computing empty states. We show the robustness of the approach and present the computational gains obtained on both the central processing unit and graphics processing unit nodes. In conclusion, our work will facilitate the exploration and evaluation of fine-tuned hybrid starting points aimed at enhancing the accuracy of MBPT calculations without involving computationally demanding self-consistency in Hedin’s equations.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Failure detection in high-performance clusters and computers using chaotic map computations

A programmable media includes a processing unit capable of independent operation in a machine that is capable of executing 10.sup.18 floating point operations per second. The processing unit is in communication with a memory element and an interconnect that couples computing nodes. The programmable media includes a logical unit configured to execute arithmetic functions, comparative functions, and/or logical functions. The processing unit is configured to detect computing component failures, memory element failures and/or interconnect failures by executing programming threads that generate one or more chaotic map trajectories. The central processing unit or graphical processing unit is configured to detect a computing component failure, memory element failure and/or an interconnect failure through an automated comparison of signal trajectories generated by the chaotic maps.

97 MATHEMATICS AND COMPUTING↗

Acceleration of the matrix multiplication of Radiance three phase daylighting simulations with parallel computing on heterogeneous hardware of personal computer

Building designers are increasingly relying on complex fenestration systems to reduce energy consumed for lighting and HVAC in low energy buildings. Radiance, a lighting simulation program, has been used to conduct daylighting simulations for complex fenestration systems. Depending on the configurations, the simulation can take hours or even days using a personal computer. This paper describes how to accelerate the matrix multiplication portion of a Radiance three-phase daylight simulation by conducting parallel computing on heterogeneous hardware of a personal computer. The algorithm was optimized and the computational part was implemented in parallel using OpenCL. The speed of new approach was evaluated using various daylighting simulation cases on a multicore central processing unit and a graphics processing unit. Based on the measurements and analysis of the time usage for the Radiance daylighting simulation, further speedups can be achieved by using fast I/O devices and storing the data in a binary format.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

Graphics processor unit with opportunistic inter-path reconvergence

A graphics processing unit and methods for comping and executing instructions with opportunistic inter-path reconvergence are provided. A graphics processing unit may access computer executable instructions mapped to code blocks of a control flow for a warp. The code blocks may include an immediate dominator block and an intermediate post dominator block. The graphics processing unit may store a first thread mask associated with the first code block. The first thread mask may include a plurality of bits indicative of the active or non-active status for the threads of the warp, respectively. The graphics processing unit may a second thread mask corresponding to an intermediate code block between the immediate dominator block and intermediate post dominator block. The graphics processing unit may execute, with threads indicated as active by the first thread mask, instructions of the intermediate code block with a first operand or a second operand depending on the second thread mask.

Kulkarni, Milind↗

Highly-scalable GPU-accelerated compressible reacting flow solver for modeling high-speed flows

Emerging supercomputing systems utilize a combination of central processing units (CPUs) and graphics processing units (GPUs) in an effort to reach exascale capabilities while minimizing the energy footprint of operating such systems. Such heterogeneous machines introduce new challenges for fluids solvers because the hardware architecture and operation of a GPU are fundamentally different from conventional CPUs. In this work, a general approach for efficient implementation of finite-volume based reacting flow solvers on such heterogeneous systems is presented. Three main challenges, namely, data access pattern, thread divergence, and thread safety, are addressed. Since compressible reacting flows require special methods to deal with chemical reactions, hyperbolic and nonlinear convection terms, and the presence of turbulence, specific algorithms that ensure GPU-based efficiency are developed. The approach is demonstrated on the widely available OpenFOAM open source software by modifying core algorithms for GPU accessibility. The scalability of the resulting solver, is demonstrated using practical test cases, including flow through a scramjet engine and the dynamics of a rotating detonation engine. Here, the solver provides near-ideal scaleup on a large number of GPUs (>3000), and extremely efficient use of the GPUs, with throughput nearly a constant even when processing a large number of control volumes.

42 ENGINEERING↗