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.
A portion of a chapter for the High Performance Parallelism Pearls 2 book. By using the NWPerf and CView software developed and deployed at PNNL administrators and users can visually see the performance, and power used by jobs running on a super computer in near-time.
Proximity to graphene has an extraordinary effect on the dynamics of entangled chains, as evidenced by strong slowing of thermally stimulated fluctuations at the surface of a thin supported film of polystyrene (PS) when a single layer of graphene is placed between the supporting substrate and the PS film. Furthermore, this slowing is due to a layer of highly viscous polymer next to the substrate being thicker for graphene (75 nm or 7.5 R g ) than for silicon (13 nm or 1.3 R g ). Molecular dynamics simulations of unentangled PS chain melts on silicon and on multiple layers of graphene show the work of adhesion with graphene is roughly four times larger than that with silicon. In addition, PS chains adjacent to silicon readily form “loops”, with phenyl rings pointing predominantly toward the substrate. Chains adjacent to graphene more readily form trains, with the phenyl rings predominantly parallel to the graphene.
The methyl ketone is a central synthetic building block for the construction of advanced heteroaryl scaffolds and systems. Reactions, including oxidative cyclization strategies, are often predicated on efficient access to this ubiquitous moiety. In the context of arenes, standard approaches leveraging Markovnikov hydration/oxidation or oxidative cleavage of the C–C π bond often afford satisfactory performance. However, when the substrate contains an electron-deficient heteroaryl core, the traditional Malaprade reaction, and related oxidative-cleavage strategies, frequently result in diminished performance over carbon-based arenes. In this work we present the development and application of an oxidative cleavage reaction of various pyridinyl isoprenes towards accessing the downstream methyl ketone for utilization in advanced cyclizations for the preparation of soft-N-donor complexant scaffolds. Further, this efficient protocol parallels the principles of Green chemistry by exchanging KMnO 4 for the toxic OsO 4 and offers the end-user an efficient, more environmentally friendly option for accessing heteroaryl methyl ketones in one hour of reaction time using potassium permanganate and sodium paraperiodate as a synergistically potent oxidative cleavage system. The wide substrate scope defined access to simple, as well as advanced heteroaryl methyl ketones. Method development, optimization, substrate scope, preliminary mechanistic observations, and a scale up reaction are delineated herein.
A scalable, stochastic algorithm evaluating the fourth-order many-body perturbation (MP4) correction to energy is proposed. Three hundred Goldstone diagrams representing the MP4 correction are computer generated and then converted into algebraic formulas expressed in terms of Green’s functions in real space and imaginary time. They are evaluated by the direct (i.e., non-Markov, non-Metropolis) Monte Carlo (MC) integration accelerated by the redundant-walker and control-variate algorithms. The resulting MC-MP4 method is efficiently parallelized and is shown to display O(n 5.3 ) size-dependence of cost, which is nearly two ranks lower than the O(n 7 ) dependence of the deterministic MP4 algorithm. Furthermore, it evaluates the MP4/aug-cc-pVDZ energy for benzene, naphthalene, phenanthrene, and corannulene with the statistical uncertainty of 10 mE h (1.1% of the total basis-set correlation energy), 38 mE h (2.6%), 110 mE h (5.5%), and 280 mE h (9.0%), respectively, after about 10 9 MC steps.
Extreme-ultraviolet to x-ray free-electron lasers (FELs) in operation for scientific applications are up to now single-user facilities. While most FELs generate around 100 photon pulses per second, FLASH at DESY can deliver almost two orders of magnitude more pulses in this time span due to its superconducting accelerator technology. This makes the facility a prime candidate to realize the next step in FELs—dividing the electron pulse trains into several FEL lines and delivering photon pulses to several users at the same time. Hence, FLASH has been extended with a second undulator line and self-amplified spontaneous emission (SASE) is demonstrated in both FELs simultaneously. Here, FLASH can now deliver MHz pulse trains to two user experiments in parallel with individually selected photon beam characteristics. First results of the capabilities of this extension are shown with emphasis on independent variation of wavelength, repetition rate, and photon pulse length.
ABSTRACT We address a simple model where the Kennicutt–Schmidt (KS) relation between the macroscopic densities of star formation rate (SFR, ρsfr) and gas (n) in galactic discs emerges from self-regulation of the SFR via supernova feedback. It arises from the physics of supernova bubbles, insensitive to the microscopic SFR recipe and not explicitly dependent on gravity. The key is that the filling factor of SFR-suppressed supernova bubbles self-regulates to a constant, f ∼ 0.5. Expressing the bubble fading radius and time in terms of n, the filling factor is $f\propto S\, n^{-s}$ with s ≃ 1.5, where S is the supernova rate density. A constant f thus refers to ρsfr ∝ n1.5, with a density-independent SFR efficiency per free-fall time ∼0.01. The self-regulation to f ∼ 0.5 and the convergence to a KS relation independent of the local SFR recipe are demonstrated in cosmological and isolated-galaxy simulations using different codes and recipes. In parallel, the spherical analysis of bubble evolution is generalized to clustered supernovae, analytically and via simulations, yielding s ≃ 1.5 ± 0.5. An analysis of photoionized bubbles about pre-supernova stars yields a range of KS slopes but the KS relation is dominated by the supernova bubbles. Superbubble blowouts may lead to an alternative self-regulation by outflows and recycling. While the model is oversimplified, its simplicity and validity in the simulations may argue that it captures the origin of the KS relation.
We describe preliminary results from a multi-objective graph matching algorithm, in the coarsening step of an aggregation-based Algebraic MultiGrid (AMG) preconditioner, for solving large and sparse linear systems of equations on high-end parallel computers. We have two objectives. First, we wish to improve the convergence behavior of the AMG method when applied to highly anisotropic problems. Second, we wish to extend the parallel package \texttt{PSCToolkit} to exploit multi-threaded parallelism at the node level on multi-core processors. Our matching proposal balances the need to simultaneously compute high weights and large cardinalities by a new formulation of the weighted matching problem combining both these objectives using a parameter $\lambda$. We compute the matching by a parallel $2/3-\varepsilon$-approximation algorithm for maximum weight matchings. Results with the new matching algorithm show that for a suitable choice of the parameter $\lambda$ we compute effective preconditioners in the presence of anisotropy, i.e., smaller solve times, setup times, iterations counts, and operator complexity.
We have simulated, for the first time, the long term evolution of the Milky Way Galaxy using 51 billion particles on the Swiss Piz Daint supercomputer with our N-body gravitational tree-code Bonsai. Herein, we describe the scientific motivation and numerical algorithms. The Milky Way model was simulated for 6 billion years, during which the bar structure and spiral arms were fully formed. This improves upon previous simulations by using 1000 times more particles, and provides a wealth of new data that can be directly compared with observations. We also report the scalability on both the Swiss Piz Daint and the US ORNL Titan. On Piz Daint the parallel efficiency of Bonsai was above 95%. The highest performance was achieved with a 242 billion particle Milky Way model using 18600 GPUs on Titan, thereby reaching a sustained GPU and application performance of 33.49 Pflops and 24.77 Pflops respectively.
The applications being developed within the U.S. Exascale Computing Project (ECP) to run on imminent Exascale computers will generate scientific results with unprecedented fidelity and record turn-around time. Many of these codes are based on particle-mesh methods and use advanced algorithms, especially dynamic load-balancing and mesh-refinement, to achieve high performance on Exascale machines. Yet, as such algorithms improve parallel application efficiency, they raise new challenges for I/O logic due to their irregular and dynamic data distributions. Thus, while the enormous data rates of Exascale simulations already challenge existing file system write strategies, the need for efficient read and processing of generated data introduces additional constraints on the data layout strategies that can be used when writing data to secondary storage. We review these I/O challenges and introduce two online data layout reorganization approaches for achieving good tradeoffs between read and write performance. We demonstrate the benefits of using these two approaches for the ECP particle-in-cell simulation WarpX, which serves as a motif for a large class of important Exascale applications. Here, we show that by understanding application I/O patterns and carefully designing data layouts we can increase read performance by more than 80 percent.
Purpose: To investigate the reproducibility and limitations of Pd-103 prostate brachytherapy using fixed length linear sources (CivaString). Methods: An LDR prostate brachytherapy case which was preplanned on MR images with prefabricated linear polymer-encapsulated Pd-103 sources (CivaString) was studied and compared with ultrasound based intra-operative planning and CT based post-implant dosimetry. We evaluated the following parameters among the three studies: prostate geometry (volume and cross sectional area), needle position and alignment deviations, and dosimetry parameters (D90). Results: The prostate volumes and axial cross sectional areas at center of prostate were measured as 41.8, 39.3 and 36.8 cc, and 14.9, 14.3, and 11.3 respectively on pre-plan MR, inter-op US, and post-implant CT studies. The deviation of prostate volumes and axial cross sectional areas measured on pre-planning MR and intra-operative US were within 5%. 17 out of 19 pre-planned needles were positioned within 5mm (the template grid size). One needle location was adjusted intra-operatively and another needle was removed due to proximity to urethra. The needle pathways were not always parallel to the trans-rectal probe due to the flexibility of CivaString. The angle of deviation was up to 10 degrees. Two pairs of needles were exchanged to better fit the length of prostate at the time of implant. This resulted in a prostate D90 of 153.8 Gy (124%) and 131.4 Gy (106.7%) for intra-op and PID respectively. Conclusion: Preplanning is a necessary part of implants performed with prefabricated linear polymer sources. However, as is often the case, there were real-time deviations from the pre-plan. Intra-op planning provides the ability conform to anatomy at the time of implant. Therefore, we propose to develop a systematic way to order extra strings of different length to provide the flexibility to perform intra-operative planning with fixed length strands.
The phase diagram of a spin–orbit-coupled two-component Bose gas includes a supersolid stripe phase, which is featuring density modulations along the direction of the spin–orbit coupling. This phase has been recently found experimentally [31]. In the present work, we characterize the superfluid behavior of the stripe phase by calculating the drag force acting on a moving impurity. Because of the gapless band structure of the excitation spectrum, the Landau critical velocity vanishes if the motion is not strictly parallel to the stripes, and energy dissipation takes place at any speed. Moreover, due to the spin–orbit coupling, the drag force can develop a component perpendicular to the velocity of the impurity. Finally, by estimating the time over which the energy dissipation occurs, we find that for slow impurities, the effects of friction are negligible on a time scale up to several seconds, which is comparable with the duration of a typical experiment.
Transformers are considered one of the most important deep learning models since 2018, in part because it establishes state-of-the-art (SOTA) records and could potentially replace existing Deep Neural Networks (DNNs). Despite the remarkable triumphs, the prolonged turnaround time of Transformer models is a widely recognized roadblock. The variety of sequence lengths imposes additional computing overhead where inputs need to be zero-padded to the maximum sentence length in the batch to accommodate the parallel computing platforms. This paper targets the field-programmable gate array (FPGA) and proposes a coherent sequence length adaptive algorithm–hardware co-design for Transformer acceleration. Particularly, we develop a hardware-friendly sparse attention operator and a length-aware hardware resource scheduling algorithm. The proposed sparse attention operator brings the complexity of attention-based models down to linear complexity and alleviates the off-chip memory traffic. The proposed length-aware resource hardware scheduling algorithm dynamically allocates the hardware resources to fill up the pipeline slots and eliminates bubbles for NLP tasks. Experiments show that our design has very small accuracy loss and has 80.2 × and 2.6 × speedup compared to CPU and GPU implementation, and 4 × higher energy efficiency than state-of-the-art GPU accelerator optimized via CUBLAS GEMM.
This paper aims to investigate the property and behavior of the hysteretic nonlinear energy sink (HNES) coupled to a half vehicle system which is a nine-degree-of-freedom, nonlinear, and semiactive suspension system in order to improve the ride comfort and increase the stability in shock mitigation by using the computer simulation method. The HNES model is a semiactive suspension device, which comprises the famous Bouc–Wen (B-W) model employed to describe the force produced by both the purely hysteretic spring and linear elastic spring of potentially negative stiffness connected in parallel, for the half vehicle system. Nine nonlinear motion equations of the half vehicle system are derived in terms of the seven displacements and the two dimensionless hysteretic variables, which are integrated numerically by employing the direct time integration method for studying both the variables of vertical displacements, velocities, accelerations, chassis pitch angle, and the ride comfort and driver safety, respectively, based on the bump and random road inputs of the pseudoexcitation method as excitation signal. Simulation results show that, compared with the HNES model and the magnetorheological (MR) model coupled to the half vehicle system, the ride comfort and stability have been evidently improved. A successful validation process has been performed, which indicated that both the ride comfort and driver safety properties of the HNES model coupled to half vehicle significantly improved.
SAND2024-11300O TChem-atm is a software library that was developed to solve complex kinetic models for atmospheric chemistry applications. TChem-atm interface employs a hierarchical parallelism design to exploit the massive parallelism available from modern computing platforms. It also supports gas atmospheric chemistry applications, e.g., the energy exascale earth system model. TChem can be used as a box model or coupled with a climate model to compute the time evolution of gas tracer species. Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.
This is a modification to 'cp' and 'mv' commands to make them multi-threaded. Simple benchmarks showed that multi-threading could reduce the time to copy a large Linux source directory by over 2x. The 'cp' and 'mv' utilities are part of the existing Coreutils (https://www.gnu.org/software/coreutils/) software package that get installed on all Linux distros. Changes: * Add '-j|--parallel ' flags to 'cp' and 'mv'. This allows the utilities to recursively copy regular files in directories in parallel. This does NOT parallelize multiple single file copies to a destination (like 'cp file2 file2 file3 dst/'). Along with this, add in new 'CP_NUM_THREADS' and 'MV_NUM_THREADS' environment variables to set the number of threads. This can be useful when you want to enable parallelism by default in /etc/profile. The maximum number of threads is internally capped to the number of CPUs. * Add a '-j' flag to 'sort' to complement its existing '--parallel' flag. This is only done for consistency with 'cp' and 'mv'. * Add test cases for the new flags. Also, run each 'cp' and 'mv' test both in single-threaded and multithreaded modes for extra coverage.
We consider the scientifically challenging and policy-relevant task of understanding the past and projecting the future dynamics of the Antarctic ice sheet. The Antarctic ice sheet has shown a highly nonlinear threshold response to past climate forcings. Triggering such a threshold response through anthropogenic greenhouse gas emissions would drive drastic and potentially fast sea level rise with important implications for coastal flood risks. Previous studies have combined information from ice sheet models and observations to calibrate model parameters. These studies have broken important new ground but have either adopted simple ice sheet models or have limited the number of parameters to allow for the use of more complex models. These limitations are largely due to the computational challenges posed by calibration as models become more computationally intensive or when the number of parameters increases. Here, we propose a method to alleviate this problem: a fast sequential Monte Carlo method that takes advantage of the massive parallelization afforded by modern high-performance computing systems. We use simulated examples to demonstrate how our sample-based approach provides accurate approximations to the posterior distributions of the calibrated parameters. The drastic reduction in computational times enables us to provide new insights into important scientific questions, for example, the impact of Pliocene era data and prior parameter information on sea level projections. These studies would be computationally prohibitive with other computational approaches for calibration such as Markov chain Monte Carlo or emulation-based methods. We also find considerable differences in the distributions of sea level projections when we account for a larger number of uncertain parameters. For example, based on the same ice sheet model and data set, the 99th percentile of the Antarctic ice sheet contribution to sea level rise in 2300 increases from 6.5 m to 13.1 m when we increase the number of calibrated parameters from three to 11. With previous calibration methods, it would be challenging to go beyond five parameters. Here, this work provides an important next step toward improving the uncertainty quantification of complex, computationally intensive and decision-relevant models.
Many scientific applications are inherently multiscale in nature. Such complex physical phenomena often require simultaneous execution and coordination of simulations spanning multiple time and length scales. This is possible by combining expensive small-scale simulations (such as molecular dynamics simulations) with larger scale simulations (such continuum limit/hydro solvers) to allow for considerably larger systems using task and data parallelism. However, the granularity of the tasks can be very large and often leads to load imbalance. Traditionally, we use approximations to streamline the computation of the more costly interactions and this introduces trade-offs between simulation cost and accuracy. In recent years, the available computational power and the advances in machine learning have made computing these scale-bridging interactions and multiscale simulations more feasible. One driving application has been plasma modeling in inertial confinement fusion (ICF), which is fundamentally multiscale in nature. This requires deep understanding of how to extrapolate microscopic information into macroscopically relevant scales. For example, in ICF one needs an accurate understanding of the connection between experimental observables and the underlying microphysics. The properties of the larger scales are often affected by the microscale behavior incorporated usually into the equations of state and ionic and electronic transport coefficients (Liboff, 1959; Rinderknecht et al., 2014; Rosenberg et al., 2015; Ross et al., 2017). Instead of incorporating this information using reliable molecular dynamics (MD) simulations, one often needs to use theoretical models, due to the inability of MD to reach engineering scales (Glosli et al., 2007; Marinak et al., 1998). One approach to resolve this issue is by coupling two MD simulations of different scales via force interpolation, e.g., the AdResS method (Krekeler et al., 2018; Nagarajan et al., 2013). Another approach, which we will pursue in the scope of this work, is by enabling scale bridging between MD simulations and meso/macro-scale models through the development and support of application programming interfaces that these different applications can interact through.
The application of peridynamics for engineering analysis requires an efficient and robust software implementation. Key elements include processing of the discretization, the proximity search for identification of pairwise interactions, evaluation of the con- stitutive model, application of a bond-damage law, and contact modeling. Additional requirements may arise from the choice of time integration scheme, for example esti- mation of the maximum stable time step for explicit schemes, and construction of the tangent stiffness matrix for many implicit approaches. This report summaries progress to date on the software implementation of the peridynamic theory of solid mechanics. Discussion is focused on parallel implementation of the meshfree discretization scheme of Silling and Askari [33] in three dimensions, although much of the discussion applies to computational peridynamics in general.