Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Supercomputer”

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 793 records · Page 44

A brief parallel I/O tutorial.

This document provides common best practices for the efficient utilization of parallel file systems for analysts and application developers. A multi-program, parallel supercomputer is able to provide effective compute power by aggregating a host of lower-power processors using a network. The idea, in general, is that one either constructs the application to distribute parts to the different nodes and processors available and then collects the result (a parallel application), or one launches a large number of small jobs, each doing similar work on different subsets (a campaign). The I/O system on these machines is usually implemented as a tightly-coupled, parallel application itself. It is providing the concept of a 'file' to the host applications. The 'file' is an addressable store of bytes and that address space is global in nature. In essence, it is providing a global address space. Beyond the simple reality that the I/O system is normally composed of a small, less capable, collection of hardware, that concept of a global address space will cause problems if not very carefully utilized. How much of a problem and the ways in which those problems manifest will be different, but that it is problem prone has been well established. Worse, the file system is a shared resource on the machine - a system service. What an application does when it uses the file system impacts all users. It is not the case that some portion of the available resource is reserved. Instead, the I/O system responds to requests by scheduling and queuing based on instantaneous demand. Using the system well contributes to the overall throughput on the machine. From a solely self-centered perspective, using it well reduces the time that the application or campaign is subject to impact by others. The developer's goal should be to accomplish I/O in a way that minimizes interaction with the I/O system, maximizes the amount of data moved per call, and provides the I/O system the most information about the I/O transfer per request.

99 GENERAL AND MISCELLANEOUS↗

An extensible operating system design for large-scale parallel machines.

Running untrusted user-level code inside an operating system kernel has been studied in the 1990's but has not really caught on. We believe the time has come to resurrect kernel extensions for operating systems that run on highly-parallel clusters and supercomputers. The reason is that the usage model for these machines differs significantly from a desktop machine or a server. In addition, vendors are starting to add features, such as floating-point accelerators, multicore processors, and reconfigurable compute elements. An operating system for such machines must be adaptable to the requirements of specific applications and provide abstractions to access next-generation hardware features, without sacrificing performance or scalability.

43 PARTICLE ACCELERATORS↗

Scalable Performance Measurement and Analysis

Concurrency levels in large-scale, distributed-memory supercomputers are rising exponentially. Modern machines may contain 100,000 or more microprocessor cores, and the largest of these, IBM's Blue Gene/L, contains over 200,000 cores. Future systems are expected to support millions of concurrent tasks. In this dissertation, we focus on efficient techniques for measuring and analyzing the performance of applications running on very large parallel machines. Tuning the performance of large-scale applications can be a subtle and time-consuming task because application developers must measure and interpret data from many independent processes. While the volume of the raw data scales linearly with the number of tasks in the running system, the number of tasks is growing exponentially, and data for even small systems quickly becomes unmanageable. Transporting performance data from so many processes over a network can perturb application performance and make measurements inaccurate, and storing such data would require a prohibitive amount of space. Moreover, even if it were stored, analyzing the data would be extremely time-consuming. In this dissertation, we present novel methods for reducing performance data volume. The first draws on multi-scale wavelet techniques from signal processing to compress systemwide, time-varying load-balance data. The second uses statistical sampling to select a small subset of running processes to generate low-volume traces. A third approach combines sampling and wavelet compression to stratify performance data adaptively at run-time and to reduce further the cost of sampled tracing. We have integrated these approaches into Libra, a toolset for scalable load-balance analysis. We present Libra and show how it can be used to analyze data from large scientific applications scalably.

99 GENERAL AND MISCELLANEOUS↗

Optimizing Performance of Combustion Chemistry Solvers on Intel's Many Integrated Core (MIC) Architectures

This work investigates novel algorithm designs and optimization techniques for restructuring chemistry integrators in zero and multidimensional combustion solvers, which can then be effectively used on the emerging generation of Intel's Many Integrated Core/Xeon Phi processors. These processors offer increased computing performance via large number of lightweight cores at relatively lower clock speeds compared to traditional processors (e.g. Intel Sandybridge/Ivybridge) used in current supercomputers. This style of processor can be productively used for chemistry integrators that form a costly part of computational combustion codes, in spite of their relatively lower clock speeds. Performance commensurate with traditional processors is achieved here through the combination of careful memory layout, exposing multiple levels of fine grain parallelism and through extensive use of vendor supported libraries (Cilk Plus and Math Kernel Libraries). Important optimization techniques for efficient memory usage and vectorization have been identified and quantified. These optimizations resulted in a factor of ~ 3 speed-up using Intel 2013 compiler and ~ 1.5 using Intel 2017 compiler for large chemical mechanisms compared to the unoptimized version on the Intel Xeon Phi. The strategies, especially with respect to memory usage and vectorization, should also be beneficial for general purpose computational fluid dynamics codes.

97 MATHEMATICS AND COMPUTING↗

LATE POP III STAR FORMATION DURING THE EPOCH OF REIONIZATION: RESULTS FROM THE RENAISSANCE SIMULATIONS

We present results on the formation of Population III (Pop III) stars at redshift 7.6 from the Renaissance Simulations, a suite of extremely high-resolution and physics-rich radiation transport hydrodynamics cosmological adaptive-mesh refinement simulations of high-redshift galaxy formation performed on the Blue Waters supercomputer. In a survey volume of about 220 comoving Mpc{sup 3}, we found 14 Pop III galaxies with recent star formation. The surprisingly late formation of Pop III stars is possible due to two factors: (i) the metal enrichment process is local and slow, leaving plenty of pristine gas to exist in the vast volume; and (ii) strong Lyman–Werner radiation from vigorous metal-enriched star formation in early galaxies suppresses Pop III formation in (“not so”) small primordial halos with mass less than ∼3 × 10{sup 7} M {sub ⊙}. We quantify the properties of these Pop III galaxies and their Pop III star formation environments. We look for analogs to the recently discovered luminous Ly α emitter CR7, which has been interpreted as a Pop III star cluster within or near a metal-enriched star-forming galaxy. We find and discuss a system similar to this in some respects, however, the Pop III star cluster is far less massive and luminous than CR7 is inferred to be.

79 ASTRONOMY AND ASTROPHYSICS↗

A new multi-energy neutrino radiation-hydrodynamics code in full general relativity and its application to the gravitational collapse of massive stars

We present a new multi-dimensional radiation-hydrodynamics code for massive stellar core-collapse in full general relativity (GR). Employing an M1 analytical closure scheme, we solve spectral neutrino transport of the radiation energy and momentum based on a truncated moment formalism. Regarding neutrino opacities, we take into account a baseline set in state-of-the-art simulations, in which inelastic neutrino–electron scattering, thermal neutrino production via pair annihilation, and nucleon–nucleon bremsstrahlung are included. While the Einstein field equations and the spatial advection terms in the radiation-hydrodynamics equations are evolved explicitly, the source terms due to neutrino–matter interactions and energy shift in the radiation moment equations are integrated implicitly by an iteration method. To verify our code, we first perform a series of standard radiation tests with analytical solutions that include the check of gravitational redshift and Doppler shift. A good agreement in these tests supports the reliability of the GR multi-energy neutrino transport scheme. We then conduct several test simulations of core-collapse, bounce, and shock stall of a 15M{sub ⊙} star in the Cartesian coordinates and make a detailed comparison with published results. Our code performs quite well to reproduce the results of full Boltzmann neutrino transport especially before bounce. In the postbounce phase, our code basically performs well, however, there are several differences that are most likely to come from the insufficient spatial resolution in our current 3D-GR models. For clarifying the resolution dependence and extending the code comparison in the late postbounce phase, we discuss that next-generation Exaflops class supercomputers are needed at least.

79 ASTRONOMY AND ASTROPHYSICS↗

WOMBAT: A Scalable and High-performance Astrophysical Magnetohydrodynamics Code

We present a new code for astrophysical magnetohydrodynamics specifically designed and optimized for high performance and scaling on modern and future supercomputers. We describe a novel hybrid OpenMP/MPI programming model that emerged from a collaboration between Cray, Inc. and the University of Minnesota. This design utilizes MPI-RMA optimized for thread scaling, which allows the code to run extremely efficiently at very high thread counts ideal for the latest generation of multi-core and many-core architectures. Such performance characteristics are needed in the era of “exascale” computing. We describe and demonstrate our high-performance design in detail with the intent that it may be used as a model for other, future astrophysical codes intended for applications demanding exceptional performance.

79 ASTRONOMY AND ASTROPHYSICS↗

High-rate Plastic Deformation of Nanocrystalline Tantalum to Large Strains: Molecular Dynamics Simulation

Recent advances in the ability to generate extremes of pressure and temperature in dynamic experiments and to probe the response of materials has motivated the need for special materials optimized for those conditions as well as a need for a much deeper understanding of the behavior of materials subjected to high pressure and/or temperature. Of particular importance is the understanding of rate effects at the extremely high rates encountered in those experiments, especially with the next generation of laser drives such as at the National Ignition Facility. Here we use large-scale molecular dynamics (MD) simulations of the high-rate deformation of nanocrystalline tantalum to investigate the processes associated with plastic deformation for strains up to 100%. We use initial atomic configurations that were produced through simulations of solidification in the work of Streitz et al [Phys. Rev. Lett. 96, (2006) 225701]. These 3D polycrystalline systems have typical grain sizes of 10-20 nm. We also study a rapidly quenched liquid (amorphous solid) tantalum. We apply a constant volume (isochoric), constant temperature (isothermal) shear deformation over a range of strain rates, and compute the resulting stress-strain curves to large strains for both uniaxial and biaxial compression. We study the rate dependence and identify plastic deformation mechanisms. The identification of the mechanisms is facilitated through a novel technique that computes the local grain orientation, returning it as a quaternion for each atom. This analysis technique is robust and fast, and has been used to compute the orientations on the fly during our parallel MD simulations on supercomputers. We find both dislocation and twinning processes are important, and they interact in the weak strain hardening in these extremely fine-grained microstructures.

36 MATERIALS SCIENCE↗

Direct Vlasov solvers with high-order spectral element method.

This paper presents the development of parallel direct Vlasov solvers using the Spectral Element Method (SEM). Instead of the standard Particle-In-Cell (PIC) approach for kinetic space plasma simulation, i.e. solving the Vlasov-Maxwell equations, the direct method has been used in this paper. There are several benefits to solve the Vlasov equation directly, such as avoiding noise associated with the finite number of particles and the capability to capture the fine structure in the plasma, etc. The most challenging part of direct Vlasov solver comes from high dimension, as the computational cost increases as N{sup 2d}, where d is the dimension of the physical space. Recently due to fast development of supercomputers, the possibility of high dimensions becomes more realistic. A significant effort has been devoted to solve the Vlasov equation in low dimensions so far, now more interests focus on higher dimensions. Different numerical methods have been tried so far, such as finite difference method, Fourier spectral method, finite volume method, etc. In this paper SEM has been successfully applied to construct these solvers. SEM has shown several advantages, such as easy interpolation due to local element structure and long time integration due to its high order accuracy. Domain decomposition in high dimensions have been used for parallelization, these include scalable parallel 1D and 2D Poisson solvers. Benchmark results have been shown and simulation results have been reported for two different cases: one dimension (1P1V), and two dimensions (2P2V) in both physical and velocity spaces.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

End-to-end experiment management in HPC

Experiment management in any domain is challenging. There is a perpetual feedback loop cycling through planning, execution, measurement, and analysis. The lifetime of a particular experiment can be limited to a single cycle although many require myriad more cycles before definite results can be obtained. Within each cycle, a large number of subexperiments may be executed in order to measure the effects of one or more independent variables. Experiment management in high performance computing (HPC) follows this general pattern but also has three unique characteristics. One, computational science applications running on large supercomputers must deal with frequent platform failures which can interrupt, perturb, or terminate running experiments. Two, these applications typically integrate in parallel using MPI as their communication medium. Three, there is typically a scheduling system (e.g. Condor, Moab, SGE, etc.) acting as a gate-keeper for the HPC resources. In this paper, we introduce LANL Experiment Management (LEM), an experimental management framework simplifying all four phases of experiment management. LEM simplifies experiment planning by allowing the user to describe their experimental goals without having to fully construct the individual parameters for each task. To simplify execution, LEM dispatches the subexperiments itself thereby freeing the user from remembering the often arcane methods for interacting with the various scheduling systems. LEM provides transducers for experiments that automatically measure and record important information about each subexperiment; these transducers can easily be extended to collect additional measurements specific to each experiment. Finally, experiment analysis is simplified by providing a general database visualization framework that allows users to quickly and easily interact with their measured data.

97 MATHEMATICS AND COMPUTING↗

Factors impacting performance of multithreaded sparse triangular solve.

As computational science applications grow more parallel with multi-core supercomputers having hundreds of thousands of computational cores, it will become increasingly difficult for solvers to scale. Our approach is to use hybrid MPI/threaded numerical algorithms to solve these systems in order to reduce the number of MPI tasks and increase the parallel efficiency of the algorithm. However, we need efficient threaded numerical kernels to run on the multi-core nodes in order to achieve good parallel efficiency. In this paper, we focus on improving the performance of a multithreaded triangular solver, an important kernel for preconditioning. We analyze three factors that affect the parallel performance of this threaded kernel and obtain good scalability on the multi-core nodes for a range of matrix sizes.

99 GENERAL AND MISCELLANEOUS↗

Cosmic Dawn: The First Star in the Universe

What was the first thing in the Universe? A black hole or a star? How did it form? Even our biggest and best telescopes cannot tell us. Direct calculation with supercomputers, however, can. The first luminous objects in the Universe were very massive stars shining one million times as brightly as our sun. They died quickly and seeded the cosmos with the chemical elements necessary for life. One star at a time, galaxies started to assemble just one hundred million years after the Big Bang, and they are still growing now. Join Dr. Abel in a fascinating journey through the early universe, where he uses the latest computer animations of early star formation, supernovae explosions and the buildup of the first galaxies.

79 ASTRONOMY AND ASTROPHYSICS↗

Parallel k-Means Clustering for Quantitative Ecoregion Delineation Using Large Data Sets

Identification of geographic ecoregions has long been of interest to environmental scientists and ecologists for identifying regions of similar ecological and environmental conditions. Such classifications are important for predicting suitable species ranges, for stratification of ecological samples, and to help prioritize habitat preservation and remediation efforts. Hargrove and Hoffman (1999, 2009) have developed geographical spatio-temporal clustering algorithms and codes and have successfully applied them to a variety of environmental science domains, including ecological regionalization; environmental monitoring network design; analysis of satellite-, airborne-, and ground-based remote sensing, and climate model-model and model-measurement intercomparison. With the advances in state-of-the-art satellite remote sensing and climate models, observations and model outputs are available at increasingly high spatial and temporal resolutions. Long time series of these high resolution datasets are extremely large in size and growing. Analysis and knowledge extraction from these large datasets are not just algorithmic and ecological problems, but also pose a complex computational problem. This paper focuses on the development of a massively parallel multivariate geographical spatio-temporal clustering code for analysis of very large datasets using tens of thousands processors on one of the fastest supercomputers in the world.

54 ENVIRONMENTAL SCIENCES↗

Self-Consistant Numerical Modeling of E-Cloud Driven Instability of a Bunch Train in the CERN SPS

The simulation package WARP-POSINST was recently upgraded for handling multiple bunches and modeling concurrently the electron cloud buildup and its effect on the beam, allowing for direct self-consistent simulation of bunch trains generating, and interacting with, electron clouds. We have used the WARP-POSINST package on massively parallel supercomputers to study the growth rate and frequency patterns in space-time of the electron cloud driven transverse instability for a proton bunch train in the CERN SPS accelerator. Results suggest that a positive feedback mechanism exists between the electron buildup and the e-cloud driven transverse instability, leading to a net increase in predicted electron density. Comparisons to selected experimental data are also given. Electron clouds have been shown to trigger fast growing instabilities on proton beams circulating in the SPS and other accelerators. So far, simulations of electron cloud buildup and their effects on beam dynamics have been performed separately. This is a consequence of the large computational cost of the combined calculation due to large space and time scale disparities between the two processes. We have presented the latest improvements of the simulation package WARP-POSINST for the simulation of self-consistent ecloud effects, including mesh refinement, and generation of electrons from gas ionization and impact at the pipe walls. We also presented simulations of two consecutive bunches interacting with electrons clouds in the SPS, which included generation of secondary electrons. The distribution of electrons in front of the first beam was initialized from a dump taken from a preceding buildup calculation using the POSINST code. In this paper, we present an extension of this work where one full batch of 72 bunches is simulated in the SPS, including the entire buildup calculation and the self-consistent interaction between the bunches and the electrons. Comparisons to experimental data are also given.

43 PARTICLE ACCELERATORS↗

Direct Numerical Modeling of E-Cloud Driven Instability of a Bunch Train in the CERN SPS

The simulation package WARP-POSINST was recently upgraded for handling multiple bunches and modeling concurrently the electron cloud buildup and its effect on the beam, allowing for direct self-consistent simulation of bunch trains generating, and interacting with, electron clouds. We have used the WARP-POSINST package on massively parallel supercomputers to study the buildup and interaction of electron clouds with a proton bunch train in the CERN SPS accelerator. Results suggest that a positive feedback mechanism exists between the electron buildup and the e-cloud driven transverse instability, leading to a net increase in predicted electron density. Electron clouds have been shown to trigger fast growing instabilities on proton beams circulating in the SPS and other accelerators. So far, simulations of electron cloud buildup and their effects on beam dynamics have been performed separately. This is a consequence of the large computational cost of the combined calculation due to large space and time scale disparities between the two processes. We have presented the latest improvements of the simulation package WARP-POSINST for the simulation of self-consistent ecloud effects, including mesh refinement, and generation of electrons from gas ionization and impact at the pipe walls. We also presented simulations of two consecutive bunches interacting with electrons clouds in the SPS, which included generation of secondary electrons. The distribution of electrons in front of the first beam was initialized from a dump taken from a preceding buildup calculation using the POSINST code. In this paper, we present an extension of this work where one full batch of 72 bunches is simulated in the SPS, including the entire buildup calculation and the self-consistent interaction between the bunches and the electrons.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

PyMercury: Interactive Python for the Mercury Monte Carlo Particle Transport Code

Monte Carlo particle transport applications are often written in low-level languages (C/C++) for optimal performance on clusters and supercomputers. However, this development approach often sacrifices straightforward usability and testing in the interest of fast application performance. To improve usability, some high-performance computing applications employ mixed-language programming with high-level and low-level languages. In this study, we consider the benefits of incorporating an interactive Python interface into a Monte Carlo application. With PyMercury, a new Python extension to the Mercury general-purpose Monte Carlo particle transport code, we improve application usability without diminishing performance. In two case studies, we illustrate how PyMercury improves usability and simplifies testing and validation in a Monte Carlo application. In short, PyMercury demonstrates the value of interactive Python for Monte Carlo particle transport applications. In the future, we expect interactive Python to play an increasingly significant role in Monte Carlo usage and testing.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Factors impacting performance of multithreaded triangular solve.

As computational science applications grow more parallel with multi-core supercomputers having hundreds of thousands of computational cores, it will become increasingly difficult for solvers to scale. Our approach is to use hybrid MPI/threaded numerical algorithms to solve these systems in order to reduce the number of MPI tasks and increase the parallel efficiency of the algorithm. However, we need efficient threaded numerical kernels to run on the multi-core nodes in order to achieve good parallel efficiency. In this paper, we focus on improving the performance of a multithreaded triangular solver, an important kernel for preconditioning. We analyze three factors that affect the parallel performance of this threaded kernel and obtain good scalability on the multi-core nodes for a range of matrix sizes.

99 GENERAL AND MISCELLANEOUS↗