Advanced Static and Dynamic Mapping of HPC Applications on Petascale Computer Systems with GPU Accelerators
Explore the source record for details and available documents.
SEARCH · Search NASA
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.
Explore the source record for details and available documents.
Not Available
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.
When simulating a lattice system near its critical temperature, local algorithms for modeling the system’s evolution can introduce very large autocorrelation times into sampled data. Here, this critical slowing down places restrictions on the analysis that can be completed in a timely manner of the behavior of systems around the critical point. Because it is often desirable to study such systems around this point, a new algorithm must be introduced. Therefore, we turn to cluster algorithms, such as the Swendsen–Wang algorithm and the Wolff clustering algorithm. They incorporate global updates which generate new lattice configurations with little correlation to previous states, even near the critical point. We look to accelerate the rate at which these algorithm are capable of running by implementing and benchmarking a parallel implementation of each algorithm designed to run on GPUs under NVIDIA’s CUDA framework. A 17 and 90 fold increase in the computational rate was, respectively, experienced when measured against the equivalent algorithm implemented in serial code.
The hybrid method combining particle-in-cell and magnetohydrodynamics can be used to study the interaction between energetic particles and global plasma modes. In this paper we introduce the M3D-C1-K code, which is developed based on the M3D-C1 finite element code solving the magnetohydrodynamics equations, with a newly developed kinetic module simulating energetic particles. The particle pushing is done using a new algorithm by applying the Boris pusher to the classical Pauli particles to simulate the slow-manifold of particle orbits, with long-term accuracy and fidelity. The particle pushing can be accelerated using GPUs with a significant speedup. The moments of the particles are calculated using the δƒ method, and are coupled into the magnetohydrodynamics simulation through pressure or current coupling schemes. Several linear simulations of magnetohydrodynamics modes driven by energetic particles have been conducted using M3D-C1-K with the δƒ method, including fishbone, toroidal Alfvén eigenmodes and reversed shear Alfvén eigenmodes. Good agreement with previous results from other eigenvalue, kinetic and hybrid codes have been achieved.
Small angle X-ray scattering (SAXS) is an experimental technique used for structural characterization of macromolecules in solution. In this work, we present BCL::SAXS--an algorithm designed to replicate SAXS profiles from rigid protein models at different levels of detail. We first show our derivation of BCL::SAXS and compare our results with the experimental scattering profile of hen egg white lysozyme. Using this protein we show how to generate SAXS profiles representing: (1) complete models, (2) models with approximated side chain coordinates, and (3) models with approximated side chain and loop region coordinates. We evaluated the ability of SAXS profiles to identify a correct protein topology from a non-redundant benchmark set of proteins. We find that complete SAXS profiles can be used to identify the correct protein by receiver operating characteristic (ROC) analysis with an area under the curve (AUC)>99%. We demonstrate how our approximation of loop coordinates between secondary structure elements improves protein recognition by SAχS for protein models without loop regions and side chains. Agreement with SAXS data is a necessary but not sufficient condition for structure determination. We conclude that experimental SAXS data can be used as a filter to exclude protein models with large structural differences from the native.
The development of NekRS, a GPU-oriented thermal-fluids simulation code based on the spectral element method (SEM) is described. For performance portability, the code is based on the open concurrent compute abstraction and leverages scalable developments in the SEM code Nek5000 and in libParanumal, which is a library of high-performance kernels for high-order discretizations and PDE-based miniapps. Critical performance sections of the Navier–Stokes time advancement are addressed. Performance results on several platforms are presented here, including scaling to 27,648 V100s on OLCF Summit, for calculations of up to 60B gridpoints.
Electron cooling is essential to achieve high luminosity in the medium energy electron ion collider (MIEC) project at Jefferson Lab. Bunched electron beam with energy above 50 MeV is used to cool coasting and/or bunched ion beams. Although the conventional electron cooling technique has been widely used, such an implementation in MEIC is very challenging. We are developing a simulation program for the electron cooling process to fulfill the need of the electron cooling system design for MEIC. The program simulates the evolution of the ion beam under the intrabeam scattering (IBS) effect and the electron cooling effect using Monte Carlo method. To accelerate the calculation, the program is developed on a GPU platform. We will present some preliminary simulation results.
Singular Binarized Neural Network based on GPU Bit Operations.
Over the past several years, the graphics processing unit (GPU) technology has rapidly gained ground in scientific computing due to its outstanding performance and programmability. GPU implementation of Monte Carlo radiation transport for dose calculations has been reported by many investigators. The majority of these studies adopted single-precision floating point format because of the higher peak floating point operations per second (FLOPS) the GPUs can deliver than double-precision. It has been known that calculation using single-precision is more prone to numerical round-off errors, especially when a single tally data is accumulated 'atomically' and repeatedly by thousands of GPU threads. To mitigate this problem, the least intrusive solution in theory is to replace the single-precision atomic-add tally function with a double-precision version. However, the complexity lies in the fact that some GPUs (Nvidia GPUs prior to the Pascal generation; all current AMD GPUs) do not readily offer such double-precision function at hardware level, and that software emulation is too slow to use if not optimized properly. This paper discusses several atomic-add tally methods with reduced numerical errors used throughout ARCHER development. The original software-based compare-and-swap method (CAS) was shown to be inefficient due to high intra-warp thread contention, whereas the improved software-based warp-aggregated method (WAG) and Kahan summation method (KAS) eliminated the thread contention and performed very well on Kepler and Maxwell GPUs, being more than 13 times faster than CAS in our tests. The hardware-based (HB) double-precision atomic-add feature available on Pascal and Volta GPUs exceeded software emulation and offered the best performance universally, so did WIB, a combination of WAG and HB methods. It was also shown that in a single-precision dose engine, KAS managed to maintain high reliability when the single-precision atomic-add tally underestimated the dose by 25%. (authors)
Starting with the Titan supercomputer (at the Oak Ridge Leadership Computing Facility, OLCF) in 2012, top supercomputers have Increasingly leveraged the performance of GPUs to support large-scale computational science. The current No. 1 machine, the 200 petaflop Summit system at OLCF, is a GPU-based machine. Accelerator-based architectures, however, add additional complexity due to node heterogeneity. To inform procurement decisions, supercomputing centers need the tools to quickly model the impact of changes of the node architectures on application performance. We present AHEAD, a profiling and modeling tool to quantify the impact of intra-node communication mechanism (e.g., PCI or NVLink) on application performance. Our experiments show average weighted relative errors of ~19% and ~23% for five CORAL-2 (a collaboration between multiple US Department of Energy, DOE, labs to procure Exascale systems) and 12 Rodinia benchmarks respectively, without running the applications on the target future node.
This paper presents the steps followed to GPU-offload parts of the core solver of EFIT-AI, an equilibrium reconstruction code suitable for tokamak experiments and burning plasmas. For this work, we will focus on the fitting procedure that consists of a Grad–Shafranov (GS) equation inverse solver that calculates equilibrium reconstructions on a grid. We will show profiling results of the original code (CPU-baseline), as well as the directives used to GPU-offload the most time-consuming function, initially to compare OpenACC and OpenMP on NVIDIA and AMD GPUs and later on to assess OpenMP performance portability on NVIDIA, AMD and Intel GPUs. We will make a performance comparison for different spatial grid sizes and show the speedup achieved on NVIDIA A100 (Perlmutter-NERSC), AMD MI250X (Frontier-OLCF) and Intel PVC GPUs (Sunspot-ALCF). Finally, we will draw some conclusions and recommendations to achieve high-performance portability for an equilibrium reconstruction code on the new HPC architectures
Abstract Not Provided
This paper presents a parallel preconditioning approach based on incomplete LU (ILU) factorizations in the framework of Domain Decomposition (DD) for general sparse linear systems. We focus on distributed memory parallel architectures, specifically, those that are equipped with graphic processing units (GPUs). In addition to block-Jacobi, we present general purpose two-level ILU Schur complement-based approaches, where different strategies are presented to solve the coarse-level reduced system. These strategies are combined with modified ILU methods in the construction of the coarse-level operator, in order to effectively remove smooth errors by targeting an algebraically smooth vector. We leverage available GPU-based sparse matrix kernels to accelerate the setup and the solve phases of the proposed ILU preconditioner. We evaluate the efficiency of the proposed methods as a smoother for algebraic multigrid (AMG) and as a preconditioner for Krylov subspace methods on challenging anisotropic diffusion problems and a collection of general sparse matrices.
Here, this paper explores strategies to transform an existing CPU-based high-performance computational fluid dynamics solver, HyPar, for compressible flow simulations on emerging exascale heterogeneous (CPU+GPU) computing platforms. The scientific motivation for developing a GPU-enhanced version of HyPar is to simulate canonical turbulent flows at the highest resolution possible on such platforms. We show that optimizing memory operations and thread blocks results in 200x speedup of computationally intensive kernels compared with a CPU core. Using multiple GPUs and CUDA-aware MPI communication, we demonstrate both strong and weak scaling of our GPU-based HyPar implementation on the NVIDIA Volta V100 GPUs. We simulate the decay of homogeneous isotropic turbulence in a triply periodic box on grids with up to 1024 3 points (5.3 billion degrees of freedom) and on up to 1,024 GPUs. We compare the wall times for CPU-only and CPU+GPU simulations. The results presented in the paper are obtained on the Summit and Lassen supercomputers at Oak Ridge and Lawrence Livermore National Laboratories, respectively.
The development of the Simplified Cloud Resolving Energy Exascale Earth System Atmosphere Model (SCREAMv1) enables global storm-resolving simulations on modern GPU-based supercomputers. However, the high computational cost of SCREAMv1 limits its routine use for process-level studies, creating a need for efficient proxy configurations. This study addresses this gap by introducing DP-SCREAMv1, a doubly periodic cloud-resolving model designed to be fully consistent with SCREAMv1 while enabling high-resolution, long-duration simulations at significantly reduced computational expense by simulating a limited doubly periodic domain rather than the entire globe. Built on a C++/Kokkos architecture, DP-SCREAMv1 achieves exceptional performance scalability on GPU systems and includes a rich library of cases for validation and scientific exploration. In this work, we demonstrate short wall-clock times at SCREAMv1's default resolution and show that DP-SCREAMv1 supports routine execution of large-domain, high-resolution experiments that were previously challenging in practice. Furthermore, we show that DP-SCREAMv1 enables routine execution of “Giga-LES” style simulations and facilitates large-domain, high-resolution simulations that were recently considered burdensome to perform. These results document an efficient, fully consistent process-level configuration for SCREAMv1 (DP-SCREAMv1) and illustrate its use for long-duration and large-domain experiments at cloud-resolving to eddy-permitting resolution.
In this paper, we report reimplementation of the core algorithms of relativistic coupled cluster theory aimed at modern heterogeneous high-performance computational infrastructures. The code is designed for parallel execution on many compute nodes with optional GPU coprocessing, accomplished via the new ExaTENSOR back end. The resulting ExaCorr module is primarily intended for calculations of molecules with one or more heavy elements, as relativistic effects on the electronic structure are included from the outset. In the current work, we thereby focus on exact two-component methods and demonstrate the accuracy and performance of the software. The module can be used as a stand-alone program requiring a set of molecular orbital coefficients as the starting point, but it is also interfaced to the DIRAC program that can be used to generate these. We therefore also briefly discuss an improvement of the parallel computing aspects of the relativistic self-consistent field algorithm of the DIRAC program.
We present the numerical methods, programming methodology, verification, and performance assessment of a non-equilibrium plasma fluid solver that can effectively utilize current and upcoming central processing and graphics processing unit (CPU+GPU) architectures, in this work. Our plasma fluid model solves the coupled conservation equations for species transport, electrostatic Poisson and electron temperature on adaptive Cartesian grids. Our solver is written using performance portable adaptive-grid/particle management library, AMReX, and is portable over widely available vendor specific GPU architectures. We present verification of our solver using method of manufactured solutions that indicate formal second order accuracy with central diffusion and fifth-order weighted-essentially-non-oscillatory (WENO) advection scheme. We also verify our solver with published literature on capacitive discharges and atmospheric pressure streamer propagation. We demonstrate the use of our solver on two 3D simulation cases: an atmospheric streamer propagation in Ar-H2 mixtures and a low pressure three-electrode radio frequency reactor. Our performance studies on three different CPU+GPU architectures indicate ~ 150-400X speed-up using AMD and NVIDIA GPUs per time step compared to a single CPU core for a 4 million cell simulation with 15 species.