Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “parallel, scaling”

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

Extending the Roofline Model for Asynchronous Many-Task Runtimes

New architectural trends and extreme scale parallelism challenge the efficient mapping of applications to large scale and emerging systems. Fine-grained Asynchronous Manytask Runtimes (AMR) offer a unique flexibility to hide the varying latencies of contentious operations, and facilitate the effective exploitation of abundant resources. By decomposing work into smaller asynchronous chunks, the impact of unpredictable operations can be mitigated through a tighter overlapping of tasks. Unfortunately, the cost of this decomposition comes at a potentially steep price in the form of runtime overhead which can at times rival the cost of the computation itself. This leads to the question, how large should a task be? A common practice for application developers is to experimentally determine the granularity of a task after a code has been parallelized. Instead, we propose a new methodology based on an extended Roofline model to provide practical upper bounds on the throughput performance of an application. First, we extend the Roofline model to support not only latency hiding analysis, but also a multidimensional amortized analysis. By combining this new methodology with a serial application and an AMR implementation, we can predict the worst case runtime overhead attribution of individual runtime features prior to the development of parallel code. Thus, this runtimecentric methodology can provide a vehicle for application/runtime codesign by providing a comprehensive bottleneck analysis based on existing runtime features.

Suetterlein, Joshua D.↗

Discrete Event Execution with One-Sided and Two-Sided GVT Algorithms on 216,000 Processor Cores

Global virtual time (GVT) computation is a key determinant of the efficiency and runtime dynamics of parallel discrete event simulations (PDES), especially on large-scale parallel platforms. Here, three execution modes of a generalized GVT computation algorithm are studied on high-performance parallel computing systems: (1) a synchronous GVT algorithm that affords ease of implementation, (2) an asynchronous GVT algorithm that is more complex to implement but can relieve blocking latencies, and (3) a variant of the asynchronous GVT algorithm to exploit one-sided communication in extant supercomputing platforms. Performance results are presented of implementations of these algorithms on up to 216,000 cores of a Cray XT5 system, exercised on a range of parameters: optimistic and conservative synchronization, fine- to medium-grained event computation, synthetic and non-synthetic applications, and different lookahead values. Performance of up to 54 billion events executed per second is registered. Detailed PDES-specific runtime metrics are presented to further the understanding of tightly-coupled discrete event dynamics on massively parallel platforms.

Perumalla, Kalyan S↗

Gyrokinetic toroidal simulations on leading multi- and manycore HPC systems

The gyrokinetic Particle-in-Cell (PIC) method is a critical computational tool enabling petascale fusion simulation re-search. In this work, we present novel multi- and manycore-centric optimizations to enhance performance of GTC, a PIC-based production code for studying plasma microtur-bulence in tokamak devices. Our optimizations encompass all six GTC sub-routines and include multi-level particle and grid decompositions designed to improve multi-node parallel scaling, particle binning for improved load balance, GPU acceleration of key subroutines, and memory-centric optimizations to improve single-node scaling and reduce memory utilization. The new hybrid MPI-OpenMP and MPI-OpenMP-CUDA GTC versions achieve up to a 2× speedup over the production Fortran code on four parallel systems - clusters based on the AMD Magny-Cours, Intel Nehalem-EP, IBM BlueGene/P, and NVIDIA Fermi architectures. Finally, strong scaling experiments provide insight into parallel scalability, memory utilization, and programmability trade-offs for large-scale gyrokinetic PIC simulations, while attaining a 1.6× speedup on 49,152 XE6 cores.

97 MATHEMATICS AND COMPUTING↗

CHALLENGES IN PARALLEL GRAPH PROCESSING

Graph algorithms are becoming increasingly important for solving many problems in scientific computing, data mining and other domains. As these problems grow in scale, parallel computing resources are required to meet their computational and memory requirements. Unfortunately, the algorithms, software, and hardware that have worked well for developing mainstream parallel scientific applications are not necessarily effective for large-scale graph problems. In this paper we present the inter-relationships between graph problems, software, and parallel hardware in the current state of the art and discuss how those issues present inherent challenges in solving large-scale graph problems. The range of these challenges suggests a research agenda for the development of scalable high-performance software for graph problems.

LUMSDAINE, ANDREW↗

Plasma Science and Innovation Center at Washington, Wisconsin, and Utah State: Final Scientific Report for the University of Wisconsin-Madison

The University of Wisconsin-Madison component of the Plasma Science and Innovation Center (PSI Center) contributed to modeling capabilities and algorithmic efficiency of the Non-Ideal Magnetohydrodynamics with Rotation (NIMROD) Code, which is widely used to model macroscopic dynamics of magnetically confined plasma. It also contributed to the understanding of direct-current (DC) injection of electrical current for initiating and sustaining plasma in three spherical torus experiments: the Helicity Injected Torus-II (HIT-II), the Pegasus Toroidal Experiment, and the National Spherical Torus Experiment (NSTX). The effort was funded through the PSI Center's cooperative agreement with the University of Washington and Utah State University over the period of March 1, 2005 - August 31, 2016. In addition to the computational and physics accomplishments, the Wisconsin effort contributed to the professional education of four graduate students and two postdoctoral research associates. The modeling for HIT-II and Pegasus was directly supported by the cooperative agreement, and contributions to the NSTX modeling were in support of work by Dr. Bickford Hooper, who was funded through a separate grant. Our primary contribution to model development is the implementation of detailed closure relations for collisional plasma. Postdoctoral associate Adam Bayliss implemented the temperature-dependent effects of Braginskii's parallel collisional ion viscosity. As a graduate student, John O'Bryan added runtime options for Braginskii's models and Ji's K2 models of thermal conduction with magnetization effects and thermal equilibration. As a postdoctoral associate, O'Bryan added the magnetization effects for ion viscosity. Another area of model development completed through the PSI-Center is the implementation of Chodura's phenomenological resistivity model. Finally, we investigated and tested linear electron parallel viscosity, leveraged by support from the Center for Extended Magnetohydrodynamic Modeling (CEMM). Work on algorithmic efficiency improved NIMROD's element-based computations. We reordered arrays and eliminated a level of looping for computations over the data points that are used for numerical integration over elements. Moreover, the reordering allows fewer and larger communication calls when using distributed-memory parallel computation, thereby avoiding a data starvation problem that limited parallel scaling over NIMROD's Fourier components for the periodic coordinate. Together with improved parallel preconditioning, work that was supported by CEMM, these developments allowed NIMROD's first scaling to over 10,000 processor cores. Another algorithm improvement supported by the PSI Center is nonlinear numerical diffusivities for implicit advection. We also developed the Stitch code to enhance the flexibility of NIMROD's preprocessing. Our simulations of HIT-II considered conditions with and without fluctuation-induced amplification of poloidal flux, but our validation efforts focused on conditions without amplification. A significant finding is that NIMROD reproduces the dependence of net plasma current as the imposed poloidal flux is varied. The modeling of Pegasus startup from localized DC injectors predicted that development of a tokamak-like configuration occurs through a sequence of current-filament merger events. Comparison of experimentally measured and numerically computed cross-power spectra enhance confidence in NIMROD's simulation of magnetic fluctuations; however, energy confinement remains an open area for further research. Our contributions to the NSTX study include adaptation of the helicity-injection boundary conditions from the HIT-II simulations and support for linear analysis and computation of 3D current-driven instabilities.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Reducing Application Runtime Variability on Jaguar XT5

Operating system (OS) noise is defined as interference generated by the OS that prevents a compute core from performing ``useful'' work. Compute node kernel daemons, network interfaces, and other OS related services are major sources of such interference. This interference on individual compute cores can vary in duration and frequency, and can cause de-synchronization (jitter) in collective communication tasks and thus results in variable (degraded) overall parallel application performance. This behavior is more observable in large-scale applications using certain types of collective communication primitives, such as MPI\_Allreduce. This paper presents our effort towards reducing the overall effect of OS noise on our large-scale parallel applications. Our tests were performed on the quad-core Jaguar, the Cray XT5 at the Oak Ridge National Laboratory Leadership Computing Facility (OLCF). At the time of these tests, Jaguar was a 1.4 PFLOPS supercomputer with 149,504 compute cores and 8 cores per node. We aggregated OS noise sources onto a single core for each node. The scientific application was then run on six of the remaining cores in each node. Our results show that we were able to improve the MPI_Allreduce performance by two orders of magnitude. We demonstrated up to a 30% boost in the performance of the Parallel Ocean Program (POP) using this technique.

97 MATHEMATICS AND COMPUTING↗

Comparative I/O workload characterization of two leadership class storage clusters. In: PDSW '15 Proceedings of the 10th Parallel Data Storage Workshop

The Oak Ridge Leadership Computing Facility (OLCF) is a leader in large-scale parallel file system development, design, deployment and continuous operation. For the last decade, the OLCF has designed and deployed two large center-wide parallel file systems. The first instantiation, Spider 1, served the Jaguar supercomputer and its predecessor, Spider 2, now serves the Titan supercomputer, among many other OLCF computational resources. The OLCF has been rigorously collecting file and storage system statistics from these Spider systems since their transition to production state. In this paper we present the collected I/O workload statistics from the Spider 2 system and compare it to the Spider 1 data. Our analysis show that the Spider 2 workload is more more write-heavy I/O compared to Spider 1 (75% vs. 60%, respectively). The data also show the OLCF storage policies such as periodic purges are effectively managing the capacity resource of Spider 2. Furthermore, due to improvements in tdm_multipath and ib_srp software, we are utilizing the Spider 2 system bandwidth and latency resources more effectively. The Spider 2 bandwidth usage statistics shows that our system is working within the design specifications. However, it is also evident that our scientific applications can be more effectively served by a burst buffer storage layer. All the data has been collected by monitoring tools developed for the Spider ecosystem. We believe the observed data set and insights will help us better design the next-generation Spider file and storage system. It will also be helpful to the larger community for building more effective large-scale file and storage systems.

Gunasekaran, Raghul↗

Performance of an MPI-only semiconductor device simulator on a quad socket/quad core InfiniBand platform.

This preliminary study considers the scaling and performance of a finite element (FE) semiconductor device simulator on a capacity cluster with 272 compute nodes based on a homogeneous multicore node architecture utilizing 16 cores. The inter-node communication backbone for this Tri-Lab Linux Capacity Cluster (TLCC) machine is comprised of an InfiniBand interconnect. The nonuniform memory access (NUMA) nodes consist of 2.2 GHz quad socket/quad core AMD Opteron processors. The performance results for this study are obtained with a FE semiconductor device simulation code (Charon) that is based on a fully-coupled Newton-Krylov solver with domain decomposition and multilevel preconditioners. Scaling and multicore performance results are presented for large-scale problems of 100+ million unknowns on up to 4096 cores. A parallel scaling comparison is also presented with the Cray XT3/4 Red Storm capability platform. The results indicate that an MPI-only programming model for utilizing the multicore nodes is reasonably efficient on all 16 cores per compute node. However, the results also indicated that the multilevel preconditioner, which is critical for large-scale capability type simulations, scales better on the Red Storm machine than the TLCC machine.

97 MATHEMATICS AND COMPUTING↗

Implementation of a cell-wise Block-Gauss-Seidel iterative method for SN transport on a hybrid parallel computer architecture

We have implemented a cell-wise, block-Gauss-Seidel (bGS) iterative algorithm, for the solution of the S{sub n} transport equations on the Roadrunner hybrid, parallel computer architecture. A compute node of this massively parallel machine comprises AMD Opteron cores that are linked to a Cell Broadband Engine{trademark} (Cell/B.E.). LAPACK routines have been ported to the Cell/B.E. in order to make use of its parallel Synergistic Processing Elements (SPEs). The bGS algorithm is based on the LU factorization and solution of a linear system that couples the fluxes for all S{sub n} angles and energy groups on a mesh cell. For every cell of a mesh that has been parallel decomposed on the higher-level Opteron processors, a linear system is transferred to the Cell/B.E. and the parallel LAPACK routines are used to compute a solution, which is then transferred back to the Opteron, where the rest of the computations for the S{sub n} transport problem take place. Compared to standard parallel machines, a hundred-fold speedup of the bGS was observed on the hybrid Roadrunner architecture. Numerical experiments with strong and weak parallel scaling demonstrate the bGS method is viable and compares favorably to full parallel sweeps (FPS) on two-dimensional, unstructured meshes when it is applied to optically thick, multi-material problems. As expected, however, it is not as efficient as FPS in optically thin problems.

97 MATHEMATICS AND COMPUTING↗

Performance of a parallel algebraic multilevel preconditioner for stabilized finite element semiconductor device modeling

In this study results are presented for the large-scale parallel performance of an algebraic multilevel preconditioner for solution of the drift-diffusion model for semiconductor devices. The preconditioner is the key numerical procedure determining the robustness, efficiency and scalability of the fully-coupled Newton-Krylov based, nonlinear solution method that is employed for this system of equations. The coupled system is comprised of a source term dominated Poisson equation for the electric potential, and two convection-diffusion-reaction type equations for the electron and hole concentration. The governing PDEs are discretized in space by a stabilized finite element method. Solution of the discrete system is obtained through a fully-implicit time integrator, a fully-coupled Newton-based nonlinear solver, and a restarted GMRES Krylov linear system solver. The algebraic multilevel preconditioner is based on an aggressive coarsening graph partitioning of the nonzero block structure of the Jacobian matrix. Representative performance results are presented for various choices of multigrid V-cycles and W-cycles and parameter variations for smoothers based on incomplete factorizations. Parallel scalability results are presented for solution of up to 10{sup 8} unknowns on 4096 processors of a Cray XT3/4 and an IBM POWER eServer system.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Towards exascale for wind energy simulations

We examine large-eddy-simulation modeling approaches and computational performance of two open-source computational fluid dynamics codes for the simulation of atmospheric boundary layer flows that are of direct relevance to wind energy production. The first code, NekRS, is a high-order, unstructured-grid, spectral element code. The second code, AMR-Wind, is a second-order, block-structured, finite-volume code with adaptive mesh refinement capabilities. The objective of this study is to co-develop these codes in order to improve model fidelity and performance for each. These features will be critical for running ABL-based applications such as wind farm analysis on advanced computing architectures. To this end, we investigate the performance of NekRS and AMR-Wind on the Oak Ridge Leadership Facility supercomputers Summit, using 4 to 800 nodes (24 to 4,800 NVIDIA V100 GPUs), and Crusher, the testbed for the Frontier exascale system, using 18 to 384 Graphics Compute Dies on AMD MI250X GPUs. We compare strong- and weak-scaling capabilities, linear solver performance, and time to solution. We also identify leading inhibitors to parallel scaling.

17 WIND ENERGY↗

Parallel-In-Time Multigrid with Adaptive Spatial Coarsening for The Linear Advection and Inviscid Burgers Equations

We apply a multigrid reduction-in-time (MGRIT) algorithm to hyperbolic partial differential equations in one spatial dimension. This study is motivated by the observation that sequential time-stepping is a computational bottleneck when attempting to implement highly concurrent algorithms; thus parallel-in-time methods are desirable. MGRIT adds parallelism by using a hierarchy of successively coarser temporal levels to accelerate the solution on the finest level. In the case of explicit time-stepping, spatial coarsening is a suitable approach to ensure that stability conditions are satisfied on all levels, and it may be useful for implicit time-stepping by producing cheaper multigrid cycles. Unfortunately, uniform spatial coarsening results in extremely slow convergence when the wave speed is near zero, even if only locally. We present an adaptive spatial coarsening strategy that addresses this issue for the variable coefficient linear advection equation and the inviscid Burgers equation using first-order explicit or implicit time-stepping methods. Serial numerical results show this method offers significant improvements over uniform coarsening and is convergent for the inviscid Burgers equation with and without shocks. Parallel scaling tests on up to 128K cores indicate that run-time improvements over serial time-stepping strategies are possible when spatial parallelism alone saturates, and that scalability is robust for oscillatory solutions which change on the scale of the grid spacing.

97 MATHEMATICS AND COMPUTING↗

Tutorial: Parallel Simulation on Supercomputers

This tutorial introduces typical hardware and software characteristics of extant and emerging supercomputing platforms, and presents issues and solutions in executing large-scale parallel discrete event simulation scenarios on such high performance computing systems. Covered topics include synchronization, model organization, example applications, and observed performance from illustrative large-scale runs.

Perumalla, Kalyan S↗

Tutorial: Parallel simulation on supercomputers

This tutorial introduces typical hardware and software characteristics of extant and emerging supercomputing platforms, and presents issues and solutions in executing large-scale parallel discrete-event simulation scenarios on such high performance computing systems. Covered topics include synchronization, model organization, example applications, and observed performance from illustrative large-scale runs.

Perumalla, Kalyan S.↗

Nonconforming Mesh Refinement for High-Order Finite Elements

In this study, we propose a general algorithm for nonconforming adaptive mesh refinement (AMR) of unstructured meshes in high-order finite element codes. Our focus is on $h$-refinement with a fixed polynomial order. The algorithm handles triangular, quadrilateral, hexahedral, and prismatic meshes of arbitrarily high-order curvature, for any order finite element space in the de Rham sequence. We present a flexible data structure for meshes with hanging nodes and a general procedure to construct the conforming interpolation operator, both in serial and in parallel. The algorithm and data structure allow anisotropic refinement of tensor product elements in two and three dimensions and support unlimited refinement ratios of adjacent elements. We report numerical experiments verifying the correctness of the algorithms and perform a parallel scaling study to show that we can adapt meshes containing billions of elements and run efficiently on 393,000 parallel tasks. Finally, we illustrate the integration of dynamic AMR into a high-order Lagrangian hydrodynamics solver.

97 MATHEMATICS AND COMPUTING↗

Domain decomposition in the GPU-accelerated Shift Monte Carlo code

The GPU solver within the Shift continuous-energy Monte Carlo neutron transport code has been extended to provide domain decomposition in addition to domain replication to enable the solution of problems with memory requirements exceeding the capacity of a single GPU. The strategy follows the Multiple Set, Overlapping Domain (MSOD) approach that is used in Shift’s CPU solver and integrates into the event-based algorithm used for Shift’s GPU solver. Furthermore, the ability to assign processors to spatial domains non-uniformly has been maintained. In this work, two different approaches for communicating particle data between domains are considered, and multiple criteria for load balancing problems have been investigated. Numerical results are presented for both fresh and depleted small modular nuclear reactor (SMR) cores. A parallel efficiency of approximately 80% was achieved with up to 16 spatial domains measured relative to full domain replication. A scaling study on the Summit supercomputer demonstrates a weak scaling parallel efficiency of over 90% on over 24000 GPUs.

97 MATHEMATICS AND COMPUTING↗

Program Correctness, Verification and Testing for Exascale (Corvette)

The goal of this project is to provide tools to assess the correctness of parallel programs written using hybrid parallelism. There is a dire lack of both theoretical and engineering know-how in the area of finding bugs in hybrid or large scale parallel programs, which our research aims to change. In the project we have demonstrated novel approaches in several areas: 1. Low overhead automated and precise detection of concurrency bugs at scale. 2. Using low overhead bug detection tools to guide speculative program transformations for performance. 3. Techniques to reduce the concurrency required to reproduce a bug using partial program restart/replay. 4. Techniques to provide reproducible execution of floating point programs. 5. Techniques for tuning the floating point precision used in codes.

97 MATHEMATICS AND COMPUTING↗

Discrete Event Modeling and Massively Parallel Execution of Epidemic Outbreak Phenomena

In complex phenomena such as epidemiological outbreaks, the intensity of inherent feedback effects and the significant role of transients in the dynamics make simulation the only effective method for proactive, reactive or post-facto analysis. The spatial scale, runtime speed, and behavioral detail needed in detailed simulations of epidemic outbreaks make it necessary to use large-scale parallel processing. Here, an optimistic parallel execution of a new discrete event formulation of a reaction-diffusion simulation model of epidemic propagation is presented to facilitate in dramatically increasing the fidelity and speed by which epidemiological simulations can be performed. Rollback support needed during optimistic parallel execution is achieved by combining reverse computation with a small amount of incremental state saving. Parallel speedup of over 5,500 and other runtime performance metrics of the system are observed with weak-scaling execution on a small (8,192-core) Blue Gene / P system, while scalability with a weak-scaling speedup of over 10,000 is demonstrated on 65,536 cores of a large Cray XT5 system. Scenarios representing large population sizes exceeding several hundreds of millions of individuals in the largest cases are successfully exercised to verify model scalability.

99 GENERAL AND MISCELLANEOUS↗